MD_MAHBUBUR_RAHMAN_NLP_Project_Support_Ticket_Categorization.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "V100",
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "eb8e213c8d3d46759dfd29c498bd4ddc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_9fdf8f36d0d3434091558433f22adf1a",
              "IPY_MODEL_5b8ca6845c5e4050aec4a1a2db4b5ca9",
              "IPY_MODEL_47cd315e497242e6bd5f7126a5dea71f"
            ],
            "layout": "IPY_MODEL_66045d939dac4e6d97aa0ee6dfc5a7ec"
          }
        },
        "9fdf8f36d0d3434091558433f22adf1a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_be53f1f4896e4cc0a1f3c0df96e87c3e",
            "placeholder": "​",
            "style": "IPY_MODEL_1ea44c1d53a24a2bbda3b2266ef05209",
            "value": "llama-2-13b-chat.Q5_K_M.gguf: 100%"
          }
        },
        "5b8ca6845c5e4050aec4a1a2db4b5ca9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_484823f23fe4464d9aa51bce9e3fe17b",
            "max": 9229924224,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_ab7eda0f0b2e4dbb9d65ba9d6173d0d9",
            "value": 9229924224
          }
        },
        "47cd315e497242e6bd5f7126a5dea71f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7116a457d0e44940a601a5ce605291e0",
            "placeholder": "​",
            "style": "IPY_MODEL_a1cf5d7e4b2e4ef9bc920de85d3e0556",
            "value": " 9.23G/9.23G [00:26&lt;00:00, 308MB/s]"
          }
        },
        "66045d939dac4e6d97aa0ee6dfc5a7ec": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "be53f1f4896e4cc0a1f3c0df96e87c3e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "1ea44c1d53a24a2bbda3b2266ef05209": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "484823f23fe4464d9aa51bce9e3fe17b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ab7eda0f0b2e4dbb9d65ba9d6173d0d9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "7116a457d0e44940a601a5ce605291e0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a1cf5d7e4b2e4ef9bc920de85d3e0556": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "## **Problem Statement**"
      ],
      "metadata": {
        "id": "82zwFmrHFb8J"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Business Context**"
      ],
      "metadata": {
        "id": "73KDwtgRdPBd"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "In today's dynamic business landscape, organizations are increasingly recognizing the pivotal role customer feedback plays in shaping the trajectory of their products and services. The ability to **swiftly and effectively respond to customer input** not only fosters enhanced customer experiences but also serves as a catalyst for growth, prolonged customer engagement, and the nurturing of lifetime value relationships.\n",
        "\n",
        "As a dedicated Product Manager or Product Analyst, staying attuned to the voice of your customers is not just a best practice; it's a strategic imperative.\n",
        "\n",
        "While your organization may be inundated with a wealth of customer-generated feedback and support tickets, your role entails much more than just processing these inputs. To make your efforts in managing customer experience and expectations truly impactful, you need a structured approach – a method that allows you to discern the most pressing issues, set priorities, and allocate resources judiciously.\n",
        "\n",
        "One of the most effective strategies at your disposal as an organization is to harness the power of automated Support Ticket Categorization - **done in the modern day using Large Language Models and Generative AI.**\n"
      ],
      "metadata": {
        "id": "-chI7K4vyr3n"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![website-6700615_1280.png](data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAABQAAAAOTCAYAAADgxq4gAADsUklEQVR42uzdeZhkdWHv/8+p6n22nn0FNMaoV70ugGK8V4MaxBX1Bq8aY0yM/uISE41GvWquP43EiBqCgICo7IuigKKCiIqKiiIqsu8IwywMs8PMdFedc/+orp4hYoShZzv9ej1POz5DTXX1qe5Tdd79XYp8+UfHpVG8KVWSpEpSBAAAAADYU3UbX5Xk+CLn/LhKiqSqHBoAAAAAqIui0/x6OiP/xD8AAAAAqJWxAX8NRwIAAAAA6ksABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKgxARAAAAAAakwABAAAAIAaEwABAAAAoMYEQAAAAACoMQEQAAAAAGpMAAQAAACAGhMAAQAAAKDGBEAAAAAAqDEBEAAAAABqTAAEAAAAgBoTAAEAAACgxgRAAAAAAKixHocAANjtVVXSbiVVmVQOBwDAg1YkaTaThgQ0mXn2AYDdX7OZzJ6bDAxtfSMLAMB/rRp747R+TbJxveMxiQmAAMDurSyToSnJl76QLL0jGRrqjAgEAOC/VjST9auT//mnyX/fP7l3Y9KwGtxkJAACALuxsdDXN5Cc8blk9QqHBADgoZoyNXnGnyQb1guAk5QACADs/prNZM6cTgDs7e+sBwgAwH+tpycZ2ZJMmWYf2Mn+reAQAAC7vSpJu935/+1WUrYdEwCA36f7lsnyKZOe/AsAAAAANSYAAgAAAECNCYAAAAAAUGMCIAAAAADUmAAIAAAAADUmAAIAAABAjQmAAAAAAFBjAiAAAAAA1JgACAAAAAA1JgACAAAAQI0JgAAAAABQYz0OAQBQd0VRpCgKBwIA2GNVVZWqqhwItosACAB4wwwAADUmAAIAtTd16tT09/dnZGTEwQAA9jiDg4O555570m63HQy2iwAIANTeqaeemkMOOSRlWabRsAQyALDnqKoqRVHkjW98Y0444YT09PSk1Wo5MDwk3gEDALXX19fXeeMj/gEAe5juOsbdPy1rwvYwAhAAqL3R0dHxP3t6vP0BAPYcZVmm2Wwa9cfD4h0wAFB72/7m3G7AAMCepDuDwXsYHtb3kUMAAAAAAPUlAAIAAABAjQmAAAAAAFBjAiAAAAAA1JgACAAAAAA1ZhdgAAB4AGWSoiyT7dx1saqq8Z0bAQB2JQEQAAAeQCNJHkbAK7YzHAIATDQBEAAAfofNI1vS19uXVA/hHxVJq9VKX2+vAwgA7BYEQAAA2Ea7XabZbOTc8y/IYZ/4dBYtmpfWaPvBv8FuNrNy1aq8/KUvzLvf/ua0Wq309HjbDQDsOt6JALD9yippVPaUAmql0ehM3b3iyl9nxcoVKYoqrVbrQf/73p7e3HX3ylx51bVJkmazmara7qUEAQAeNgEQgO1QJe120jM2va2qHBKgNrpr9910862ZM2d2pkwZTNl6kJuBVFWaPT2ZV87O2nXr0mq109PTTLtdpdlUAAGAXcOQDQAevLLshL+imcxfkjSbncWuGk0REKidG2++LT3NZlqtdlpVlXZZ/t6PVlVltNVKb29flt21PMuXr+ieQB1QAGCXEQABeBCqZHQ06etL5s5PhmcmF52TvOWVybmndv6u3XaYgD1ee+yXGUuXLcvmzZvT09vbmb77IP99kaSqqvT39WblPfdk2VgArPySBADYhQRAAH637oi/RjNZvFfSP5Rc/LXkdQclH3lncuetyVdOTlbfnQxO6dweYE8/7yW5/oabU5btNLZj4b6qqtJoNFI0GrnxltuSbaYVAwDsCtYABOCBrl474W9wMJk2MxndnJxzWvLlk5Pbbth6u2ZPsmFdcuG5ySv/Ollxr98tARN0Gqp2STTrjtS77oabMzraTrPZTPshjnAuiiJlWWZoYCDX3XBj53TZbO6i41imKJyXAWCyEwAB2KosO/Gvrz9ZsChZuzb5+pnJSUclK5eNXdmO/U9VJeXYRfFJn05e9MqkbyBpjdrqEthuVdWJVo3GrolW3c974y23pErncbRarYccI6uqysDAQK6+/oZdejyLopGyrFIURiECwGQmAAKQVGXSLpPBoWTGzOTeDckZn02+cmpy1+3bXhmPTY+rtl6pJ51RgN/4UvKqNyRL70h6vLwAD1273Rlxt6tGrHWn7ibJ8uUrM9g/mLIstyuclWWZgf6+3H77Hbvk6+g+5ipVGo1i/DHtqrDK7mV7v69h0r9l3uZ1AvY0vnMBJvcVQGeqb29/snjvpKySL30hefWByVEf7cS/zrCRrbf/rVeSsZeSM45LNqxNBgatBQg85AuqqqrGp8n+9Oe/yIEvPjQf/tinOv99J22iUY6du9auW5dlK1ZmYKBv/O+26412o5HNIyO5/fY7x7/OnXM8O3+e+ZVz89RnPj9f/ur5448nSdptG5JM+ovARiNFUfjw4eMhfoh/7MkM0QCYrOGvLJMpU5Np05MNG5JTjkrOPilZtfL+ce/3XfyWZScQrlqZfOtryctfmyy/w++YgAel3S7TbHbOF7+5Y2k+eeSx+eZFF6d/oD8nn/6lPP5xj8n/OuRFOymcdcLYypWrsmzFysyfO+dhRbuenmbKdpXrbrwp++yzJGVZ7vC1ANtVlWajyBW/vCof+dgRKYoi7/nnw/LFr3wt7/77t2S/pzwpzWaRdrtMo1EYBTaJdNeDHB0dzV0rVqWvv98v7OChvH3u/CRl0bw5nbVevdtlDyMAAkyqdy5ja/wNDCYz5ySrViSnHJOcflxy78bObboXg1X10C8MTjk6OfgVSe/A1vUBAR5Au12l2SzSbHbW2DvimONz6lnnZPOmzZk3f156ms1sHtqc/3vY4XnkPnvnqU9+4k7Y0KJz/rv9zrty770b07NwQUZHR7f/3opGWq3RXHvDjXn+8/5kh48ALMsyzUYjdy1fkbe96/+kv68vM2ZMS1lWufqaG/IXb3x7XnTQc/JP73xb5s2ZPf5vuiNbqPtbgCLNZrJ0+d357FnnZc7wtLRLo0HhwWikyGh7NAN9A3nDqw/J1KGhFGW5dSYM7AEEQIDJ8a6/s87flOmdEX+r70mO/0Ty1dOStau3eXfT2L7RAN2L2ruXJ9/+avKS/50svzNp9jr2wP20qyrNokiz2QlO557/zRx57Odzx2/uzJy5szNzeHparXba7Xb6+/szMjKSt77zfTnviydl3pzZO3Qdu+5IxOtuuHF8/b/t3Y24u05UkSK33Hrb2Cl2x10odj9fVZV589+/J+s3rM/sWbPSanV+GTNnzqyMjo7maxd+O9/9wY/yhte9Om9+w+u2mRbc3mU7FbNzNXua6Wk009ffn1ar7HZv4HeeYJNGo0g1UqS3t7n1XOkXJ+xhBECAOuuO+BuamgzPTO66Iznzs8kZxyejI/d/87I9I/621ehJylZy6jHJn760syPw6KjfjAJjp5gqSZXm2Ai+n//yynz834/OFb+8MtOmT8vixQtTVtV4sMrY6LRp06blntVr8tdveUe+esYX0mg2tzvK/T7d+7zu+hvTP9h/v7/bnvsqyzL9A/1ZtqKztEIn0E38Y9/2Pt/2rvfnmutvyOJFC9Nubz2W7XY7jUYjC+bNzZaRkXzi08fmnPMvyD/+3Zty8POek+bYcX04XzN7jqLodL+GpxoexA+MnxfqwVUZQB2VZdJuJVOnJguXJJs3Jkd+OPnz5yYnH9WJf40iKRqd8DcR09LKVueK4q47ku98PRmeMzH3C9TglNSdZtrIipUr897/+9G89m/+Lldfe0MWLVqY6dOmpD022u6B/u2cWTNzw0235C3v/D+da7Gi2KHTaa+5/qYM9venepjro5VlmcGB/ixfuTL3rFkz/ncTfm06FuwOP+KYfPOi72XRwgX3i3//+TH19fZmr8WLsmr16rz1ne/LX/3tP+TKq68ZnwpcWhcOAGpHAASo11V2Z1ffKdOSxfska9cmRx+WHPqs5KzPJSNbxsJf0dnxt9pBF3mnHJOMbkl6TAEGto58+8wJJ+Ulh/5lvnzeNzJ71nDmzBlOWZa/d1fadllmwby5ueh7l+TfPnlksgNGqXWD4sb77ss9q+9JT09fyocZGauqSn9/f5YvW5WVY6MAJ75bdu7w7HO+nuNOPDWLF87/vQGvqqrO6MopU7J40aJcdsUv8pq/fnP++aOHZ/3GjXa5BIA6vh9zCABqoGx3wt/U6cmivTpr/B3+/uTVB3am+3YXKR4Pfzto5Ez3fu+8LbnkwmTm7KQ96vmByXpqGgtR3/7u9/PclxyaT336uBSNIksWL0xRFL83/N3/9FJl0YIFOfbE0/Klr5y/wx7rDTfclLIs09MzMW+Te3p6smnLptx6+53dr2QCT7lVkiI//fkV+eBH/y1zZs18SGG0qqpUVZV5c+ZkxvQZOevs83Lgwa/IyWee/Z8+BwCwpxMAAfboq+t2J+5NG04W7Z3cvawT/v78wOScUzoj/MbDX7lzpuQ2xpaXPeWozjTknv4kppPBZNSNUb/69dX59dXX5RF7L0l/f39ardZ239+CeXPzfw/7eH582c+TCQxU3Xu57sZb0hpbL29CTtNlmSmDg7nuxpuSZMI22uhOq166bHne+o/vz+DgQPr6+rbreHTXB9xrr8VZtWZNvnvJpeN/bz1AAKgHARBgT1S2k/ZY+FuwOFn2m+Sj70j+4qDk3FPHzvDNnRv+xh9bq7NS8m03JT/8VjJrdtI2ggQmo+46dK9+5Suy15KFuXfTlocV7KqqSm9vb4aGBvMP7/lgli1fPnFr1o09rhtuvGl8R9yJOHN1pgEP5NobbkwmaOpyd8ffLVu25E1/94/ZtHlLZkyf/rBj6JYtWzJlaEre9Nev9c0LADUjAALsSboj/qbPTBaOhb8PvyN5/QuTb3x57MzeHfHX3oWbcIy9vJz8mU6o7LHpPExG3dFuixbMz1Oe+MSsXbv2YY+Aq6oq06ZNy72bN+UNb3t3RkZG02g0HlYErKoqPWPnqd/csbTz/ydoI4yqqjIw0D8eACfi/rbd8ffmW27PnJnD2z2qcvysXRTZvHlL5s+ZlWfsv+/9nj8AYM8nAALsCdrtznTe6bOS+UuSO25JPvR3nfD3rXO6V9q7ZsTfA16hjl0433xt8pPvJsOzOl8DMKkURTEepg563p+kVbYmZLReq9XKnFmzcuMtt+Zt//i+zpvasY1GtkdZlkmVbN68KStW3p2hwcG0qyoTMfm1M2qxJ+vXbsjatevGTunt7b6vbvz76OH/ke9874eZv2Bu2hNwTBvNZtasXZuXvujg8c8FANSHAAiwO+uO+Js5O5m/OLntxuQDb0re8JLk218bO5OPjfhrt3d9+LvfK8zYqL8vHNl5fD1GksBk1B1F9oqXvDDTpk7NyMjIhNxvu93OovnzcvH3fpiPHH5Eks7qA78vXFVVlXa7SrvdHt8Eo9lsJkVy36aR3LViRQb6+yc0gDUajbSrMrcvvWv8mHQ/d7vdHn8sv9fYbc44+9x8/pTTs3Dh/JQTtMRCq9VKT29PDj7owLHja+1WAKgTARBgd9SNeTNmddb4u+Ga5D1vTN50SGd33exmI/4eSDk2He3GqzujAGfOsSMwTEJFUaSqkt7enhz4P56R9Rs2TuhGGIsWLciJp56V07/0laQofucae+12O61WK0VRpNks0mw2U2xz+7uWLc85X/tGtmzekp6e3gkNgEVRpKfZk/O+dkFuv+OO8b/rPJbm+GNptcq0Wq0H/NxlWaZoNPLjyy7Ph/71k5k3d96EbVTS09PMmjXr8pQnPiGP+6NHd15i/NIGAGrFokwAu4tuyGsUyaw5SW9f8qufJqd+Jrnskq23azQ6t9sTptQ2Gp2v6+RjkgOekzR7Pc8wCbXb7fT0NPOyl7wgZ3/1mxO6s2xRFJk3Z3Y+/G9H5A8esU8O2H/f8YBWlmUajcZ4aOtav2FDbrz5tvzwx5fl8p//MncuW55779uU0ZGRDA/PSFlO7Pm1qqpMGRrKued/M+dfcFEGBvuzYO7c7PfkJ+WPn7F/nvC4x2R4eEZ6ehrjv5+vqmr88Xc3/bj1ttvzd+/+QKZNGUp/X++ETP3N2MDC0dZIDnzWM5PuaEBrtwJArXhlB9g9ro47G2VMH+6Ev19clpz2meSn3996m2azE9PKPWhaVvexXvPL5PIfJk85IFm9qvO1AJNG90f+Gfvvl8f+0R9m5cqVmT59+navhbetqqrS39eXoaHB/N27P5DzTj8xixbNH/u8W881V197fb7/o8vys5//Ir/69TVZvXpNBgb609vbm76+3vT29qZ/6pQdeAwamTp1KKOjrdx37+Zcv/bm/Pra63LM50/O1ClDecJ/e0ye9tSn5KlPeUKe/cxnpCga44+/KIqMjo7mLe98X0ZGRzJzeHjC4l+SjIyMpL9/IIe+/MW/ddwAgHoQAAF2le6Iv2YzmT2vM/LvV5clJx2dXPHjrbcrGkmqPXcTjUazs5bhyUclT3tWZ1QgMKkURTNlWabZ08wzn7Z/Tj7ri5kxY/qE3X+7LDNtypSsXr0mf/22f8jXzz41zUYzV117Xc457xu55EeXZdU9q7Nu3bpMGRrK9OnTss8+eyXbrBnYXZNvR+pMhe4Ex2RwfCRkq9XKNdfdmMsuvyJDgwOZPTw7++37pBz6spfkgKfvmyLJ377jvbn51t9k0YJ5Exr/Go1G1q5bnwP2e2pmDg+PPV+Fb1oAqBkBEGBXaI8mPX2d3XEbjeSyHySnH5v88rKtt2k2t+7+uyfrXqheeXly+aXJk5+e3LPcdGCYZLpx7RUve0HO+PJXMjramtDQVJZlZs2amaVLV+Q1b3hrehs9ufQnP83A4GAGBvszZWgwM4dnjE+tbe+iX6p0QmM6v9gZ0yiKzJo5nNmzZqbdaufezZty4cXfy1fO+3oe//jHZuH8ufnxTy7PgvlzJzT+ZWylhnXr1+cVh7yo8/LUbhsBCAA1JAAC7CzbjvibuzhpjXSm+J706eSaX2293Z4+4u+3L3e3rgV46meS/Z7Zme5c+ZaAyaQblR7/2MfkUY94RJYuX5FpU6eknMCgVZZlhoen58abbk1VVtl77yWdTUjKMtXYSLvd9Cx5vyA5ZXAw04aGMmf2rKxYsSpL71qWOXNmT/gIxaIosnHj5vzBI/bJs5/5dN+kAFBj5mEB7Ayt0U74mzO/sxvu9y9I3v6q5D1/szX+dXdcrHbTXX0f3lV5588rftTZ2GTqcI0CJ/BgdQPW/zrkRVmzevWE7WL7n02fNjUzZkwbH+lX7oTpvRN9nNplmbIsMzQ0kOEZM3bI4282Glm3YV2evv9TMzQ01JmmbfQfANSSAAiwo5Tl1si1YK9k+ozkB99K3vDi5J/fujX8FY0kRdKqeRBrjA06P+moZHAoscYUTMLTYueXAc97zrMyPDycLVu27JD15trt9i6b4jvRqqraYV9Lq91OUTTy8hcdPP65AICaXo45BAATfrnWGfHX25vMW5jMnJ1868vJmw/thL9brh87A48FsarMpJgPW41dwF7+w+SKnyTDc4wChEmm2WymSrJk0cLs+5T/nrVr1xlxtgtt2rQlSxYtzAFP23f8+QEA6kkABJgo3RF/jWayaJ+kfzD51rnJa/80+Zd3JTdc1bldd7RL2Zpcx6eqtu4AfObxSX9fZ+djYFLpjmZ7/p8emLKsz0i9PU1PT09Wr12bl734oLFTtNF/AFBnAiDAw1VVSauV9Pcn8xcmU6Yn55yS/O3Lko/+Y7L09rEzbs/W209W3bUAf/Sd5Nork6kzktLFP0wmPWOjzA55wfMzdeq0jIyMOCi7QKvVykB/Xw5+7nOSJO126aAAQI0JgADbqzvir6c3Wbx30uxJzj09ec2fJJ/8QHLbzUkxiUf8/c5XnrEpZicdlUydnpRGncBk1NfXm+c865lZt369qac7WbPZzJo1a/KUJz0hj/7DPxj7O5cFAFDryzCHAOAhqsrOiL+BgWT+4qSvLzn9uM7mHp/6YHL3irEzbKOztJ9pVb99/JLk0ouSa65IZlkLECab1tjP/Mtf8oJs2TLa2QiJnWp0tJXnPvt/dJ6PVmuHbMYCAOw+BECAB6s74q+3rzPir6ySs7+QvPJZydGHJcvu7Iz2Gx/xZzrVA7rfWoAnJP0D1gKESaY5dg54+n5PyWP/6FFZu95mIDvTyMhIBocG82eHvKTzfDj2AFB7AiDA71N2R/wNJQsWd6b6nnRU8vqDkyM/nGxY1xm90mh04pYRfw/umCbJxecnN183thagYAqTRVEUabfbaTSbOWD/fXPvhnuzJwxAK4oiRVGkURRpNBppNhppNBppNIrx/7a7azabWbt+ffZ76pMyffrU8a8LAKg3ARDgd+mO+Osf7Iz4a21OTjk6+bP/kRx/eLJ61TYj/ioB6yG/Ao29BJ1ydDJdAITJ6g2ve02mTZuaVmvXLQXQ3QG3KIo0GkWazWaazeZY3Ot8JMloq5VNW7Zk472bsnHjxqzbsDEbN27Mxo2bsmnTloyOjo6d3rb+u2az0bm/RmO3CW2jIyP5/17/F+NfEwBQfz0OAcB/Upadj6lTk2nDyT13J8cenpxzcrJx/diNis60VdHqYVxxj/158fnJX7w5WfSIZN2axFQ02Ona7XYajeZOHYXXnXa6YePGVEnKshwPbTtTJ/j1pqqqjIyMZNPmLRkZHUnZLlOW7bTbVcqqnZ5mT6ZPm5YZ06amf3AwzWYj/b29GR1tpVW2smXzlmzYsDF3r7onI6OjnfBXNNJoNNNoFunt7c1Af3/6+/vTaDTSbrdT7uTXkKIoUlVlhgan5K5ly5I8Ob09Pamq7NTnvt1um3YMADuZAAjQVZad6btDU5Lh2clddyRfOik5/Zhky0j36qnzZ1XZvfbhqsrOjsBlO/niicn/OTzZsMZxgV2gG2OqqkpZdgbo7qjRalVVjd/3hRdfkvd88CPp7e1Js9nYKSsoFGPTd5MqZVnl3ns3Z+36tRkdbWX2rFmZM2s4c2bPzGMf85g85lF/kCVLFmWfvRdn9sxZ6enpSW/v73773Gq30xodzbr1G3LbHUtzxx135uZbb8vV192Yu+++O/fcsya3r7wzRZEMz5iRKUNDnfUQiyJlWY6PRNxhp92qSlUlg4P9ee+HDsuKVffkTa9/bYri/s/Ljvi8ndjZSLNZiH8AsAsIgABl2YlRU6Z3pqIuuyM59ejknNOSzZu6V4ydUX+VEX8Te+zHpvx940vJn/9tZ0fg++5NrEcFO9WNN9+aRz/qkSmKYnwQbqvVSrPZnNAotG1kOuXMs/Phj/17Zs2ckf7+/h0evzpfWzMjIyNZu25dRkZG02q386hHPiIvfeFBecLjH5vHP/aP8phHP2r731g3m+lpNjMwMJD58+bm6fs++X7//dbbf5Nrr78xV151XS67/PJcfe0NKRqN9PX0ZurUoQwMDKQqy7R38MjAvr6+zBwezsc+dVTuvHNZPvyBd4+NDqwm/Plut9vp6em5X/S7+dbb86hH7vNb3xMAwI4jAAKTV3fE39TpyYyZyW9u6qxHd+YJW0PftiP+YsTfDtFodJ6L049L3v+JZMP6pMfLE+xo3WmYS5cuy8te81dZvGBhDnnRQTnkRS/IksUL0jP2c1glKdvtNB7mGnbbhp5jPntiPv4fx2ThgvljU1B3zPm1GFuPr0qyefPm3HPPmkyZOpRH/8Ejc+CznpkXv+CgPGLvJQ/w8lClGv+FTyPbMzO50/DKsZeSzqjDR+6zdx65z9554UHPTZKsWHF3LvzO9/LNb30nN95yW+667fbMHB7OjGlTUzQaO2xUYFVV6enpyeKFC3PKWWfn3vvuzScP+1CKokhZVmk0Ht7z3J3OXRTF+PfRPatX5+sXXpzzzr8gv/jV1Tn1c0fmj5/+NAEQAHYSV1jA5FO2O1e006YnU6Ylv7mls6nHeadtXdOv0exEQDv67jxf/2Ly6jclc+Z3dlY2RQx2mKqqxkdkffzIz2TLyEjWrF2TIz/zuXz6uM/nj5++Xw78n8/MC5//vMyeNTx+27IsU1ZVeh7iz+e2keeTn/5MjjruxOy1cGGKZmOHxb9ms5l2q5XVa9blvs2bMm/OrLzpr1+bFz7/eXnC4x7zW4+v3W6Ph7pOAGs+zM+f+91HN4xVVZVGs5lGUWT+/Ll53asPzetefWhuvf03+d4PfpSvfeNb+dVV12ZwcCDTp01Nf39/Wjtgo47ucd97yeKce/6Fufe++3LsER9Po7F9IwHb7XZSFGk2Glu/X9rtfOOi7+aCb38337/0J7nvvk2ZNn1qBgcH8okjj81XTntaJ9CKgACwwwmAwORRllvD37Th5Nbrk2M/npx76tbbNBpj6/u1Ha+d+bx0RwF++ZTknR/qBEBgh+kGl59d8ct841sXZ9G8eSkajUyZMiXtdjs/vfwX+d4Pfpyjjv989nvKE/O//+wVefp+T05/X3+6g+HaVZU8iPUCt407h//7MTn6cydlyaIFKRo7Jv41iyJlqqxctSplWeVJj39cXv/aV+VPD/yf46PRMr7xSeer2Xak2o7SnYK87XHJWFRtNpvjowP/6rWvyk9++vN89qTT89PLr8iq1Wsyb87s9PT07JARgWVZZsnihbnw4u/njW9/Vz575Cc6IwHb7TT+i9BbVVW6gxzvt65fVeWKK6/OmWefkx//9OdZcfeqNFJk9uyZmTVzuPP4pya/uuqanPnl8/Kq/3WIAAgAO4EACNTf2KiETJuRTJma3HpD8ul/6Yw46+qO+LOr767RPe7nnJS85o3J0NRk833WAoQdoKqq8fD1sU9+OkMDg2kURVplme5P3Jw5s1MURe677758/9Kf5Zvf+l722XtJnnfgs/OcP3lmnrH/vmkWxfgAt9+1XuC2YefIYz+XYz53UvZavHD8v02koihSFEXuWbM2W7ZszgFP2y9vecNf5oCn7Xu/x9MNbrt6I4rucek+jna7PT4194Cn7ZsDnrZvfn3NtTn1zC/nq9/4VhpFMjw8nN7e3s5ouwk9BZd5xJLFufh7l+bt735/jjz8o2k0mw8Y5u63rl+2DnK8+dbb8/ULLsqFF1+Sa6+7PgNDg5kyOJiF8+em0Wim1Wrd72scnjEjRx//hbz8xQenv7/fzsAAsIMJgEB9le3OClAzhjtB6fqrk7NOSL51ztbbGPG3++iOAjzrhOQdH0mWbrAWIOwA3ajzxa98Nb+66uosXrQo7W3iX7rTOZP09/dnYGAgs2YN59777svnTz09p33xK/mDR+yVFz3/eXnpCw/O4kXzH3hkXXfzpCSfO/n0HHH08Vm0aMH4ZhMTpkiaRSMbNm7MmnXrc8B+T80/vPVNedo2G3C0t1nDcHeNTP95J+Zms8gT/9vj8m8f/kD+5nV/nn8/5rh855JL09fbk5kzZ47t6Dtxx7FVtrPX4oU5/8LvZGDgX/Lxj3zgfs/VA63rt2bNunzz2xfnnPO+mZtv/03WrV2XGcPTs9deS1IUnbUUOx9bpzAXRZF2u53p06bljjuX5lOfPi7ve9fb/8vRhgDAw+fKCqifdjtpFMn0mZ0Rf9ddmZxxfPLtr229jRF/u6/zTkle+fpk6szk3g3WAoQJPT22x3fCPfq4L2TGjBm/d/puNwANDAxk7yVLMjo6mqXLludTRx+fI475bJ6+/1PywoOelxcd9LxMmzblfiGrKIqcf8FF+egnjszCBfPTmOD412w20xodzZ13r8wj9lqSD77nnXnZiw8e/++tVuu3dqDd3d1vJ+Z2Oz3NZh79h4/MMZ/6WH7y05/n3444Jr++6prMnj0zAwMDKSfwdawsy+y1eGG+dO5XM2v2zLz3H9762yMVy3a+eeF38p3vX5rv/uBH2bBhQ6ZOGcrg4GD23ntJqrL9oHYxbrVamTd3bk770jl5zStfkX32XmIUIADsQAIgUB/dEX8zZyeDQ8nVv0xOPSb5/oVbb2PE3278/JWdcDvaTs47M3nz+zoBEJgw3bjyyU8fm6XLl2evJYsf9AYTVVWl1WqlURSZPm1aZkyfnna7nSt+8etc+pPLc+RnTsh/f+J/y2sPfUX23/ep6e/vzS+uvCrv/ed/ydw5syd86mqz0ciadWszsmlLXveqQ/P+f3p7mo2t02kbjcYOX9dvh79R32Z6cLPZzAFP2zfnnP65fOHUM3LU8V/IqlWrM3fOzLTLTNhO9WVZZsmiRTnhC6dk78WL8ppDX54kuera63PW2eflkkt/nGUrVqQoisyZNSszpk8bD8UPdbOS/r7etNut/Osn/yPH/sfh4h8A7Mj3FQ4BsMdrtzthb8asTvj71c+SL34uueSCba96O4HJiL/dWzl2AXvGccmf/WUyNJRs3mwtQJiQU2WVZrPIb+5cmjO+eG7mzZ27XbvLVkmqbc6lc+bMTpLcu2lTLv3Rz/KtizrrBb7w+c/JBRddkr6+vvT39U1Y/CuKIlWSO+9alkc/+lH58Pvflf2f8uSxr7FMo1HULiR1v55Wq0xPTyN/9dpX57nPflbe96HD8uOf/jwL5s9NX19P2u1qwo7xrJmz8u9HHZ/bfnNnfnnlVfnZz3+ZwYGBDE4ZzOIFnU1c2u3Ww3peW+125syek+98/0e55NKf5NnPPGB8qjEAMLG8ugJ7prLshL+yTGbPTeYuSG64Onn3XyVvPXRr/OuuQ9Vud0b+sYc8v1Vy9knJ8ByjNWGCNJudkH7Y4UdmpD2a/r7eCbnfdruddrudwf7+zJo1nL32Wpx777svnz3pjGzYuCFTp06dsGm/jUYjWzZvzvLlK/PKl7803zz71PH4V1VVms1GrXeT7elpjI+223uvxTntc0fn3X//5qxdtz7r1m1Iszkxb+2rqsrAQH96enpy8mln5Yabb8neey3OvHmzM2VwMO2yTKvVmpCX1aJIpgwN5uNHHD3+HAMAE88rLLDnaY92RvTNmpPMX5z88rKt4e/H3+1e6WZsBXLhb497ZRp7afrKycnKOzvrOLZFQHg4uuvE/eDSn+Q73/tB5s6Zk9YE/1x1d9gtyzKDAwNZvHBBBgcHJiz+NRuNrNuwIZu2bMknD/tgDvvQ+zovCWOjGItJMlK4u9txOfb8vflv/jInHXtkpk2bmpV3r5qw0Y+d3XqbWbhwQYanT09Zlmm3ywnfvbmqqsyYPj3XXX9TTjzlzPt9vwIAE3iZ5RAAe8jV69govjKZuzCZOSe58vLk71+VvPN1yWWXjF0ZGfFXi+e6aCab7ku++qVk2rBjAg/3Dd9YWP/YEUdnytSh7OhUVo6tB1eWExT/ms2sWLkqs2YN56yTjstLX9jZ6KOqqjQn6W7hjWazE11TZf99n5RzzzgpT3r847N0WWd9vol6jlut1oSu3fi7vl9mDs/IcSednrXr16fRaOzwzwkAk+69g0MA7PbGR/zNTeYvSn76/eQdr+18XPGT7tVh58/KiL9aqMZGf5xxbLJudTLQ73mF7f1xGvvZOfH0L+a662/KjOnTJ3wU1w59s9po5M67lueJT3xszj7xs3ncHz16fIRYMcnXBy2KIo0UabfbmTVzRs448TN56cHPy513LU+1h32PTps2NavuXpUjjvrs2Mu6DUEAYELfUzkEwG6pO+IvSeYuToZnJT/7QfLGlyX/9IbkF2Phb9sRf9TI2KXr5k3JOacls+ZZCxC2Q7vdTlEU+c3SpTnu8ydn1uyZKfe0+Ld0WQ5+7oH54onHZ9as4VQ2ifgtzWZzPIp+8l8/lDe+/s+z9K7lKctyj4mkrVYr8+fPzZfO/Wp+ceWvO68EfvEDABP3vsohAHa/q4DRpLe3s7nHnHnJDy5I3v7q5L1/k1z7q+7VTudPI/5q/Ao19hL1pc8n99ydDE6xizNspyuvvi433Xx7ehqNFMWe8favURS548678tIXPT9HffJfkrG14Qrx73ecMhvjwex973xb3vX2v83yFXfvMVNpu2sbrrz7nlx08feTsV2dAYAJeq/gEAC7hW1H/C3YK5kyLfnhRcnrX5h88K3Jr38+doVgxN+k+p5oNJL1a5MLvpxMHxZ74SHqTqN88UHPzfFHfjwb792YdevXpbmbR7RGo5E77lqeV778xfn3f/3Q2CnByL/fpyiK8Qj4lje+Pu/++7dk+YqVKctqtx4J2Gg0snnLSJavXJn//wP/lH96x1uTJD09pgEDwETpcQiAXa7V6qzxNn1mJ/Bc8OXk7BOTG67a5uqgJylbW9eGY3LoBr9Tj05e8qqkr7+zJmQKxwYeope/5AX5w0c+Im9553uzYtXqzJ87e7ccHdYoiixdtiwvfv5z87EPf2DsVFCJfw9SNwIWRZG/fcNfZMvIlhxxzAlZsmjBbvl4G41GNmzYkNHRVo782Efyguc/d/w5n+xrPALAhL7mOgTALtEd8ddoJAsWJwNDyYXnJK/90+Swd22Nf90Rf2XLMZuMqqoT+9avT84/K5k1J2kZ/ckDGJs+uO1Huh/brCXWblcpy3ba7c5HVVWTYp2xquocgyc+4XH58mmfzx/94SOzbPnyrcdqN9FsNnPXihV51h8fkCMP32barxD0EH8cto4E/Ps3/03+8s8PzbLlK3e749hsNrPqntWZOmVKTv3cUXnB8587/r3qOQeAiSUAAjtZ2Rnx19efzF+YTJuenH9m8qaXJ//6T8nS28fOTmOnJ2v80Ri7CDzrhGTD+mRw0FqAk9E2LaBoNH47XI2FvG0/0v1IMmVoKEnS39+bRqOZZrPz0b2fsqoyOtpKaywK1u7wFZ0o1G63M2/u7HzllBPyvAOfnTvvWrbbxJZGo/H/2DvvOLnKev+/n+ecM31ntrd0EhIIgopSBRWx4L2ACAgYBa7tqmDD8pOiBr1iBQW5iFgoERALXMFCkRa6CBZACKTXTdlstu/OzDnP8/vjzEw2IZDdzZYZ+L5fGUImk9kzTz3fz3ye75f2jg72nbs3P7n0e+GOIcd+96DPt4uAF577Bd7+tjezaXM7jlMe7amVYt2GNubO2Yubr/8Fr33NfoWjylLdWRAEQRDGAzkCLAjCxGAKQl4kCg310NMJf/wtXHsZbG4bEuCrsACsCDzC0LGjNLRvhrtvhRNOh01r5TusV1H/K2uxxpSKQ1s/v9OLFI7r4kY8YvEknuPieBFcz8NxXbq7uln86OPkrINSluaGJhob60kkYiQTCTzPQyuF9rbfFllrMYWfrwuC4yuBYrVYx3W44pJv8d1Lf8xPr1lEQ10d0Wi0VEl2Mq6rt7eXVCLJVZd+l0jEE/FvDFBKldrxyh98h+NP+SCr1mygtrZ60vq6KEyu3dDG8f/5Ti799jcKc076WxAEQRDGExEABUEY9+AdYyCRCHP89ffBb66B310Nbeu2v07rgugnbj9hVxFjYWhcdwUcczJEYlII5pXe5SbsX+t5WC8SVoG2Bg1UNTSRSiRJpqtJVqWJxuJEYwm8SBRlC6uItYXVxBD4hj/cdT83/d+fw2O/gcFYSzKZoLmxgRnTpzJj2lSmTZ3CvvvM5cADXkM0GsFhe8VxYwzG2FdEUQKtNUEQ4DgOX/7cWcyZNYOvXfQ9cr5PVTI54cKQAnK5HN09vVx95Q9oaWkuXZ8wNv1dEgEv+x7vOe3D9PX3k4jHJ9ztqrUmm83S3rGNsz56Jl/8zCdL80vEP0EQBEEYX0QAFARhfCg6/mIxqGmAbVvg5mtDAadzayHqKzhqrBXHn7D78QSwdRPceTOc8AHYsB482cZekf3sONhUGtwI9GwjsWEtdY8tpn7mbFLTZ+Kmq3G1iwGMnycwhsD3yedyL/nWVckkmaqqHZ7zfZ/unj7+9vd/svihx8j7Ppl0FZl0milTmnnTIW/kXW97K3vvPRutdSkzQRBYtK7sY4qO45SO/p70nv+ktaWJc768kK3btlFfWzuhxUG0Umze3M4XP/tJjjj0kNL1CWPYxlpjrWVqayvfWngen/rCucRj8Qm/hp6eXrLZLN/9xlc46fj/KNwCSIEXQRAEQZgIJHISBGHsA3hjIJGEqmro3gaLLg+r+na0D40ERPQTRhg9FipB33AVvOu9YeXoIC9HgV8BKGPC0/+xeJgmwPdJL3+eumXP0rjsOao2b8A6HkEqiXHi5LNZsmZgh+UE9MuKCNbaF4laSini8SjJZJyG+vDf5vN5stksS5a8wD+feobLfvxz9pk7myMPP5QjDz+UNx16EI6zXfirZKfaUAHzsIPfyO9/dTUf+8yXWLJ0Ga3NzRPiBNRas2lLO4cf+kY+8ZEzSn0lOeDGr7/fdfRb+MApJ3P9b37H1JYWggnoZ8dx2NK+lUwmzVWXfY+D3vC6XY5DQRAEQRDGDxEABUEYO6yFaBzqGsK8fjddBdf/BHq6inf5218n4p8wUowfjqGN6+DeP8O7T4bN68ARAbBy+9SAUthECrwIetsWWh+7n9an/0Zy21ac7AB+PMFAdS0W0NaWRN+xcgwVhUHfD8eXBmKxGPF4vJCrzLBxUzs/veYGrrvxt8zZayYnH38sJxx/DJl0uuSkM4YdhMHKW74tzc1N3HLD1XzqC+fzl/seYEpr8w6FJMaDbDZLIpHgkosWlq5DBKHx7WelFF/78uf465P/YMPGjdRkMuPm+CwW2dmwcSP7zN2bn/3oYpoa6wGLtSL+CYIgCMJEIlGTIAhjR7o6DM6v+h6c8U648rsF8U+FFp0hFTkFYZTRZPj7dZdDPgtuVNqkEikU9ghzg2aIb1zH3n+8kTf97PvM+8utJDq3ko/G6K9twI/GUdaG4t843roopUqpJkNBzxAEAcZYUqkkM6ZPpa62hrUb2vif71/KUf9xMl+76GJWrl6DUqok/lVqBeFisQjPc7nqR9/jI2e8n7aNm8nnfRw9Xm0euv++9JlP0NBQj7VGBKEJ6mftOHzjK/+Pgf7BUPwep59lrWXdhjaOOvJwfrfoZzQ11hfERiV9LQiCIAgTjAiAgiDsOUEAqSp48E444ZBQnOnuDMUaVQipxfEnjAXFcdS2Fu77c+g2DfLSLpUkQAQBxBLYdDVeRztzb1nEIddexozHFqOMZaC6Bj+WCN14QVAWXxoYY/B9H6UUmaoqpk5pIRqNcONvbuakBR/hCxd8ndVr1oafryCwVKIQWMwTB3DBlz7Lhed/gc7OTnr6+sY8R5vjOGzt6OCIQw/itJNPKI4OmSAT1M8AB73+AE476Xg2b2nHHev+1ZpcLkfbxs18aMFpXHXZ90uVnSW/oyAIgiBM0j2ANIEgCGOCBZqnDo3uxPEnjNPOVche8csfF1yAEWmTCkAVj/tW1+H1djHnjzdx+M8uZtq//orxIvTX1mMcN3T6lfG6UXQGep7HtKlTiMRj/PGOv/AfJ5/ORd+/jN7eXrTW4350dtz6ach1f/DUk/jp5RejlKJjW+eYCjf5fB5rLQvP+0LYrkEgjrAJpHjk99xzPk1DfR09/f1j1v5aa3r7B+js6eZbF36Zr3z5s6WfKcU+BEEQBGESwyhpAkEQ9hjHga5tcODh8K6TwudE+BPGC1M4rrZmOTx4N9SKC7DssQabTEEsTtPfFnPI1T9k5mOLMa7HQFU1xnFCtx+Vs25Ya/F9n4jr0tLURCaT5urrf80xJ36ARTf+FnYS0yqJodd95OGHcPMvf86U5ibaNm4ak/d3Xc3m9nZOPfE9zNlrVpjfUVxhE7xtOxhjSCTjnP3RM9m2rWtMBECtNVs7tqGBn192Cae89z2l+SLOP0EQBEGYXEQAFARhjFYTDV1b4QP/Hf654PYRhHEbbwDX/AisLy7AcqW4DqRribet53W/vJL9b7sRnc/RX1OLdZwhuf0mhmJRgrFyOxWLiDhaM31qM9lsnoXfupgPffJz9BdcVZUqAlJwPM6YPpWbb7yat735TfT09e1R2yml6OsbpK6mms+d9ZHwZ4xTAQphd8touI4uOPUk5s3di+6e3j3rW63o6+1j9qyZ3Pqrazji8INfNJ4EQRAEQZjEvV+aQBCEMYokoLcHZs2Fd7yneMcv7SKMD8VcgKuXwgN3Q019mItSKBuUDcKq4IkUrQ/fzcGLLqN29QsMpGswkSh6jPOCGmNKD9/3C4+AINj+8H2ffD5PNpsln89jxvgafN8Qj0eZMX0qix9+lE9+7tzCUli5a6HWmiAISCbinPfFz6AAsweCplKKjs5tnPLe48lkMgSB5ISb3KU0nANnfexDdHV1ofXox6rrOHR2dXPO2R+lpbUZ35c1WRAEQRDK6r5OmkAQhLGLJCwMDMKpHy5GFiICCuO4gxW2sOuvAGvAcaVNykdVwKaq0bl+9rvp5+x7x2+x2mUwlQnLPOyhI64o9BWFPWstkUgEz/OIRaNk0mmqq6upqanZ5aOxsZG6ujoikQjGgO/7pfcK3xvUKAtSFB2BM6ZOZfEjf+Wa628qXXOlYm3YFr+77c/09/Xj7oFgl8vlqK+p4ayP/ldhGsut6GRSFKePfdfbed0Br2HbHuR6DAJDLBbjj7ffDYWj3oIgCIIglFH4JE0gCMKY4TjQ2Q7zD4S3vKsYXUi7CONDUVB54Vl45D6orZNcgOWABTK1JNYs5+BrLqNlyT/Jpmswrou2Zg+62xD4flhFVGvi8TiZTIb6+nqam5qpr6+nvqGB2vp6qqurSWcyVFWlSKW2P6qqqshkMqX/b2ioo6mpgalTp9Lc3Ew8niAajaI15PzcqEU7W2yGqhS//+Md4Q3XkAq7FdWd1paEnIcefoxoNDbq93Jdh83tWznphGOJxWOAlS1iklFK4RfG+QdOeS/9A4PYUY57YwypVIJ7H3iIbDYPVObxd0EQBEF4pSICoCAIY7yqKOjpgg+eXYwIpE2EcRxvhW3sl1eE/+9ILsDJVRM0pKupeepxDrrxSmLdnQxkakb9drbg8jO+j+tFSWUy1NfV0djQQF1dHVVVVcRiMZQuCA3WYo3BDwKCIa6+oY9sNlt6GEPJ7eS6LnV1tbS0hGJiXX0diXi85A4ciZChCmJIMpli9dr1rFq9FoCgAtfDYrXYZ559jmf+/TzVNZnScyMaGkA2myNdleL973svFI5MC5OPW1hHT3rPsUxpbaF/YHDU7xWJROgfGORPd929w/gRBEEQBKEMQidpAkEQxnZVcUIBcN4BcOQ7Cs/JUiOME0VBZclT8OhiqK6VXICT1Q9aQ6qKKQ/czut/dy1GKXLxJGoUope1odsPrUmkUtS1tlLfUEdNOk00FgPtDDmua0YkzmmtSw8KDjdrbem9BgYGcRyHdCpNdU0tU6ZMIZPJjMLJFDrnBgYGGchmi09VHEWB9MHHnsAPfJxRWvYc12Hrtm285YjDmNraUnAWSu6/cqE4vhe87wTaO7aN+pi34zgYa1j80MOl+SYIgiAIQpmE6tIEgiCMfSQRQG4A3v/x7eKAnPMSxm0nG+ICdJ1QhBYmDmPAdSGRZMa9t7HvXb8nn0xhveiIK/zawjFfrV3SNTU0NDVTV1tLzHFQ1hIURLrxVNKKgkU2m0Up8DwXa+0oCnkocvk8mXSKTDoVPlOBy2Dxc99zz2KqM+lRFwDJ5300ivedcGxh2Ij7r5woOvWOPeYdVGeqyOZyo1sOgoB0VRXPLlnK4OBgqYiMIAiCIAhlEDZJEwiCMOY4HnRuhQMOgoOPpGIjX6EyKAoSzzwJf30QauslF+CEURD/4klm/uVW9r7vzwxUZbCOM7JCH9YS+D5Ka1I1NdQ3NZFOp/Gwpbx/E00k4hLk82zY0EZXV9fIb7C0pre7j733mkVrc3PpucqaWmEftnds44UVK4nFoqPqC60Vg4NZpk2bwpsOPTjcJqTyb1nhumERpdaWZg4/+A1s6+wclUPTWEsykWDp8lU89e/ndhhHgiAIgiBMLiIACoIwTpEjMNAHZ362EBWI20MYr7Fmt7sAb/p5WIzGlYrA448Jc/4lU8y8+/fsvfgOBtLVKK1HJP4ZYwisJV5VRWNLCzXpNC4Gk8sxGauG1hrX8ejq6qZt42Z838d13RE7AI0x9A70cfqC90HBYaUq7IuQonPrwYcfI5fP43nuKNvUYVtnF8e/+52FKSuCUDni+z4A73jbWwvVrEc3Ax3HwXE0jz/+j9KfBUEQBEGYfEQAFARhfHAc6O6E/Q+Eg44oRoHSLsLYM1RUefIheOw+iCdFdJ6IW4h0Na0P3c2cB+5iMF2NYvgCVygwBGHxjYYG6mprcYqVfu3krBee55HP59m4MXT9aUeN2LUXinyKNWvWsuB97+WYtx9VWBIrVwR54u9PEeR9HMcZ1eFrPwjwXIe3vTncC0YrLAnjvW2HY/TYY95OKpkiN8pjwEEQUJVM8sgTTwyZE4IgCIIglMHduyAIwjhhfPDzsEByAQpjj1IKVTxqWhT7jj4Oahqg4GQRxglrIJ2h4YlH2PeuW8hWZbAjEMqstdjAkkinaWhuJh6JYHyfyZKFtNZoR9PV1cWmTZvI+3kcZ+SuP0dr+voHWN/WxodOP41vLTyv9HkrroutLR0LfWHFcuKJONaM/HM4jkN3Vzfz589j/r5zC8/J7We5rqkUjgMfcvCB9Pb3j+rYujGGeDzGc0uWbs//J65PQRAEQZh05A5MEITxw/Ggox3eeCS89uBChCHLjrDnQarSuiAiFYLLd5wA190JX78cZuwF2aw4Tser/a2FVIbU8ufZ7083kYunwpx/wxUHggCUorqhnrrqanShmu9k4WqNMYbNmzazbdu2HSoEj2hMKkXbxs0k4zH+9+JvsfC8L0JBSKtEB1SxT9Zv2MjKlWtJJuIExjDiUigKBgYGOGC/fQHwjRFHWBlTFOze/uY3093dO+q8la7rksvneOLv/wrfVxzZgiAIgjDpSHQkCML4Yi3kBuG/PlOM/qVNhFGhlMJ1w4qs1hiqgMix74Or70BdeBk0t0DbeujuEqfpeGEMNhZD93Sy/603YNEEnjdsd08QBGjPo66piWQsGh73nawbIK2JRCJ09/ayYcMG8vn8qHL9ua6mb2CADRs38e53vY0//PaX/Mc731ZY/mzFil1F1+L6jW1s3rKFWCw6KidjEAR4nsdbjjh8+54glH2/H/i6/ampzpDPj66gkuM4BH7A089KIRBBEARBKBdEABQEYXxxHOjcBgceCvu/sRh5S7sIw9+otEYphbUW3w/zkH3mc5+jO5/jul9cATUN2I3roKcHHC3jazzxPFCa/W+7gUTnVvx4Aj0C8S8ai9HQ2EhUa4zxJ/FjeBjfZ9OmjXR0dOyR62/jxnainsel37mQS7/zDTLpqpJ7rpLEP2ttKS9jENiSTvfckuU4rkYpPSrtLp/3cSMebzr0oHBLkPlZ5lt26OadMX0Kc2bPpKe3d8T5K621aK0JjOXZ55eWngvHVkBQGGsiCgqCIAjCBMdV0gSCIIw7Jg+BgdM/Wfiz5AIUhrFBaY3jOBhjSgHlBRdcQGdnJ5f98IfgepxWG6Mp3wHGGfVRNWF4KGsgmWL6Q3fS8Py/w6Ifdnj+vSAIiMXj1NbX4ypVEMj0pI2p3t5eNm7cSH//AJ7njVio01ozODDIuvVtvPXIw/njbxZx7DHbK9yW+1gsijG+75PP+yWnolKqUMFVlSr+Ll2xgmQ8gTHBiJdtR2uyg3lmz5xJxPPCcSRrf3nP89L8hL1mzKCvf3DE36kU3yMWj7J5y1YoiO7h2HJwCmNNKYXv+/i+TxAEIggKgiAIwjjjShMIgjDuOB5s2wqHHQX7vhae+xcoB6wUahBejC7k9ysGoZ7nce655/LVr34VryAihC4SRcSt4or5jZz8j05IReV44XhhDDaZJrFyOXs9fA/ZVBXKjEz8q6urR2EnLd+f67oEJmBr+xb6+gZwXBd3hGJUUbzavHkLmeoM3/3mV3jfe44tNJEpuVXLgaKYYgr9B7r4qyT07Uwul2Pjps08+/wy/v38Cyxbupx/Pv1vqqpSI67ca8MGo2+gj4Pf+LrSNYkAWAHT3Vo08PrXvoY/3P4XgmB0R7+rkkmWLlvB6R//FPPnzWP+vLnsv+9cGhrqqaqqKs3LXY5bUxq9JUFdxo4gCIIg7OH9sDSBIAgTE1EEYUXgMz8F534srBAsCEMoBnlFgSiRSHD++efzxS9+kWg0WgoqKRxTKx49PWnuXkxb+jfW+uC4EIgGOKYoY7CeB/kc+935O5RvCGLesI7+Fo/91tbVoQiYDO1Pa00QBHR3d9HT00vgByUheaTvMzA4SMfWbbz9rUfyja/+Pxob6guf0056ZduiaF4UUIrCikMhFcNOrNvQxtLlK1m5ag0rV61lxarVrF67lg1tm9COwnM9XM8lmUzijfAIKIAqtFn/wAD777tPaTzsLPgI5UfxmPb+8/fF9RyCIBiV+FYUmZ96ZglPPvk0vp8n7+epSlUxY/o0Zs2azuxZM9lr5nT2mj6d2XvNJBaLsn3I7jjujAkIjC0VgnJEEBQEQRCEESF3YYIgTFBE4cC2DjjkKJi7H7zw77AisJXKgK92dnb8pVIpzj//fM4999xS0GkKlUOHupaUUgTW4ngpvj+7jtOe2kbgxoreI2GsUAqSKabe92eq166kv7oOPYx5a4MA1/Ooqa9HlRw9E73sOGSzg3R2dpLN5nEcjeM6I/z4BddfezuJeJyvf+VLfOCUE8PPaA0KheOoMuimXbv6unt6aGvbzFPPPMfzy5by7PNLadu4mYH+frp6eunt6ycajZKMR4nF40yb0oIqnvm0do+qt/q+TywaYdbMaTKPKoiiU3OfubNxtLNHwq3jOFSn04Xj46o0Lja0tbFsxUpu678DN+JRlUySSiWpqckwc/p0XjN/Hq+Zvw/z5uxFuiqN57lo7UiKV0EQBEHYA0QAFARh4ii4tzj9bPjqWaH4p5Qc23yVsrPjr6GhgXPPPZfPf/7zQ4bMdsffLoPLwu+n7rc357zwKG0+aAdEVh4blDHYeAJ3Uxt7PXZvePR3GAKrMQblONQ2NOIO6eOJHltdXdvo6ekHLK7rjOp9BgcHad/awRGHH8a3Fn6ZKS3NpbHpjMIZN150dnazet06lq9czZo161i9dj1r1qzlhRUr6OntI+K5uK6H4zpEPQ/P82ior6O1uQlb6CNjDMba7Wv1nowdpcjmcjQ3NdHY2FB6TqictdnzPBob6uns7MSLeFgzur165/mvlKIqlSJTVYXSmsAY8vk82WyWNWvWsWLFGu66+37y+TzadZg+pZWZM6cxZ9YsZs+cwaxZM5g5fSoN9XWFHyBZzQVBEARhOIgAKAjCxOEo6NwCR7wd9toHViyRXICv0uByqOOvrq6OhQsX8ulPf3qHgPGlHE07RZL4Flwd53/3beSkv7djU3ERlccIqxR4EfZ+8A68/n4G0jW7df8Vj6BmamuJaDXh4p/juGQHB+js6iaXyxby8o28wi8F118yFmfhuV/gjAXv2+HzlYP4VxQh777vAb54wddRWtPd3Yvv+8QTMeKxOJl0mrra2mLvFH+VqrBmc7kdPvNYzvNsLkdTfT211ZnSc0JlMWvGNB7bvIVkIkEwhu7qwBh8AN8fUnzGIR6P7+AWNMbQ3dfHX//2D+67/2EslpqaDD19/Xz1S5/ljPefQqAMjiiAgiAIgrBbRAAUBGEC0eAH4e9nfAou/JTkAnw19f5Owl9zczMLFy7kE5/4ROk1+Xwe13VHJBQ4BfvHifNmMXNpO6tyYd0ZyQW4ZygTYFMZkitfoPnpv5NNpYd19NcYQzKTIRWPE/j+hI4vgO7uLrq7u8Ox4YzO9ZfNZtnS3sGbDn0j3/zaeUyf2hqKFmXm+hs6b3p7+2hpaSaTTuNojcVgjCUITMlJu8t+HgdXXlgJGfLZPLW1NWjtFH+YTKwKY9qUFh54OAcq9P6OZQ+qIWOiKKzvqhJwPBolGY+jG+oLx5Mtm7d0MDAwOPSdBEEQBEHYfTQuCIIwgTgOdLbDke+E6bMlKHyFo5RCa43WulSgoKWlhUWLFtHW1lYS/4p/53neiAUJpTS+teAkuXjvesgNEEhAuMdYrUFpZj16X1i9dxiirAl8vEiMdDo94eJfPp+nvX0zXV1dpXE30rGqtKa9vYOBbJYvf+FsFv30cqZPbS255cpR/APwHJdoLIbjOPi+TzaXI5cLRly5d6xvMfN+jurqNAC+L7OykggKQtyU1lbyfj7MvTdJzmprLUEQkMvlyOfzKAWxWJRIJCIdJQiCIAgjujsTBEGYaHwfXB1WBA5v76VNXoE4jlNy/BljmDNnDosWLWLDhg2cfvrpUHAuhW4hvUdOpKKd/aR95zAlYcGXDW5PUMZAPEly1VIalv6bXCK126q/1lrQDtV1GRwbTNi1aq3p6elhy5YthUIfzojHktYOuVyedevbeO3+8/m/G37Bf5/5QSi4/pRSZZ2/zhQKkhSPzoePyb8oYy3pqqrCKi/rfEVRTNFQU00+8MsiZAjnoMLa4pdGku1VEARBEEZ0zytNIAjCJNzFQ8dWePO7YOqsUP/TkpHgldG1qiTmFY8dzpo1i1tuuYWlS5eWhL8gCEbt+HupMeVbQMW4cp9GGMiG+euE0c9Rx2Hm4/cX3H+7n58mCEhVpYl40Qmr+KtQdHZuo6urqyQkj3S8OlrT3tFB/0Af/++zn+TGq3/MXjNnlL3rr7zHDxgMjnKorakuPiVUEMWjuNXVGYwflOMQk1ElCIIgCCNEBEBBECZh5dHg5yESgw8U8r9JLsCKx3XdkuPPWssBBxzALbfcwooVK3jve98LQxx/o3Fp7Q6nUPv3uLmz2CttsPmw7owwwsDaBNhEkuj6NdQvX4IfS6B3U1fZGoMTiVCVSqEm6OhvWGRigJ6eXhzHGbH4px0HP59n7YYNzJ87h5t/+XM+8ZEzwuVoiJNOGMUYKsozSpGuSkqDVDCpZFKENkEQBEF4pYTh0gSCIEwKjoZtW+Dtx0JTayFqlCCj4gL9IbnW/ILws88++3Drrbfyr3/9qyT8jbnjb5fXUswFmOCHcxskF+AosShwXFqe/Qfe4AC+G9lt7i9jLVXpNFopzIRdJ/T09I3a9dfRsY3u3l4+d/bH+O0vf8bec2aXXH9SrXYM+sdaUBbPK+Zpk7lYkVu1q0MXrDGyRwuCIAhChSN3uIIgTBIK8nmIxuDMT2+P6IWKYajjD+Cwww7jtttu47nnnuP444+HcXb87fKaCoPo+H3nMDNhwTey0Y0IC14E+vuY8tQT5OJJ9G7y+ZmCsBuPxzFmYo4KKqUIfD8sNDKCcVUskrFuwwZmz5zBb679KZ/+74+En0Ncf+PST57nlZZ8oQLXecfFdV2MlQ1aEARBECodiYsEQZjM6BC2bYOjj4XGllB8EOdN2Qf0Ozv+DjzwQO69914eeeQRjjvuOJggx9+uL1CHuQCJcsW+zZILcKTNZyzEEtQvfYZYTxemKN68DNb3SaYzOJPVzsP8uVprOju76Ozs4qyPfpjf33QN8/edK64/udEUXn6CleaZrKSCIAiCIPdlgiAIo1yBNGT7IVEFCz4ePmekql+54nneDo6/o48+mrvvvpsnn3ySo446CgqOPwpOq8lyUhVzAf7H3FnMyWhsziBlHEbSgA5Nzz+NxWLUy7ecMQYnGiURj2GCSSgUoNSw1gwLbN7STktzE7+69id8/tP/DYDvl6frz1qL7/ulR973S0UZKgVVcIdmc/ntnSBUHH7g4/sBaCpuDAqCIAiCsFP4LU0gCMKk4rrQsRnefRLU1G0P6oXyCOKHOP6K4t6RRx7J448/zt13383RRx8NOzn+Jv+aC7kAdYwfzK2HfJ5AxtTu280YbDSG3rqR9Po1+JHY7ot/WEs8kcDVEyuxWjv8ZcLRiu6uHo5525HcfssNvPY186Hg+nPd8roNCgV2i1IK13VLD891UUphDFhbIV+SKFBW4ZeEYRGPKpHADwiCAC0hgyAIgiBUfugtTSAIwqRHibkc1DTAGWfDZd+QOLFM8DyvlMMP4JhjjuFrX/sahx12WOk1vu/jum6YJL6cNreCOHTcPrOYvWQLy/MG7U5cgYrKxEI0Rs0Lz5Ds2ER/TQP6Zd11FqU1iWQS4+cn/Fodx0Vrxe6Mh1ZpfD/Hmw97IwDZXI5oJFJ2xxmNMWitUQp6evu574GHWL12PcYEtLQ08o63vIWamkxFfHcbLhkKg6W3r1emVqXuzUD/YD8YI8d/BUEQBOEVgAiAgiCUQZyhoGsrvOMEWPS/sK0jPB4sx4EnoStUwWlkSo6/Y445hosvvpj99tuvENzbkljhumW6jVjwFbjEufQ10znu8Q3geWBlTL1kk2kF1lCzejlG775fjbFEY3EijjMpx3+VUjiOSz7vv/zrgMBYtnV1A+Do8nP9UchRCHDZj3/GTbfcxpYtWxjM5tFK4bia79ddyX8tOIWz//tDO/y7cp2AWoM1ho5tnTK5KnRbBtjW0YV2HfleThAEQRBeAYifXxCEMliJNAwMhEeAT/lYUV2QdpnQYE+9qKrviSeeyBNPPMHtt99eEv983y8IL055VkstVqwtJqxf/zTHrrmTFq8PY91SfkDhJW4J/IDaVS/gR2O7cf+BDQKiyeSkjlmtdWm87ubF5P3y6/uhlYf//dzznHDah7j8J7/A932mTpnCvL1ns/fsWcycNg2tFN/54RV89ZvfL33+XYmA5SMMaizQVRBehcrblgG2dnaGDu+y2ZPtkP8KgiAIgjDCu31BEIQywHFg6xZ4zwJIpEpBuzC+FEWUYtEBgJNOOolly5Zx880384Y3vAFjTCnHX9k6/orBabFoxfrncP5yOcGDN0DHUr7hbQA0gWx7L91+kQhex2YSnR0Yd/e5HB3PI+ZMrlN3uAU8rLXk87myae6dKw9f/pNfcOp/fZzlK1cxdUor8ViMIAjIZrNkczn8ICAajbLXzGlce8NN/P6Pt5feh4LLUStdyBno4DgOjtal9rHFnznBYyrqRejsDgXA4hcMQmWFCG1tm4hEIpjCfJvYObLjPuU4TmF8a6naLQiCIAijQI4AC4JQHigF2UGoqYXTz4arvittMq7NHbr4/CHVRU8//XTOPfdc5s+fX3pdMcdf2WKDUPQrBoPrl8Bz90D7anCjOMkMYPmoWct57EW7TaDJYUQI3HE8YLFelJo1K1DGgHZe1mNjfB83mcSNRjG53Pb2n0CKx9BfdngUP5+1DObyZdHWQ6/7+WXL+MrXv8vfn3qaxoYGopEIwZDj1EMFF2stCk1jQx0/+snVnHDsu0tFdwbzOTZtbieXz+FotyCUaCKRCJ7nEfE8tNYExuzg8h3P9cUU8oh2dMgR4ErdIwDWrV9P1ItMmHhcFPuKc8T38+RyOXK5HPl8ELrQNWzcuJnBwcHCpLLgyBeGgiAIgrA7RAAUBKGcIg7o6gorAt/4E+jpklyA4xBcFXP8FR1/CxYs4Dvf+Q7Tpk0rCRRFd1JZO/603u742/ACPPsXaF8DrgfJmvB5a/FRuCrHJXYZZ3IgkJeBsDMW8CKkt2xABz5GK7R56ZDfKkVUaxzAn0Qnzm5dQNaiAct2h+ukNfFOuf5+8vPr+PHPFxFYw7TWVoy1O4h/uxz21hCPxdnQ1sZTTz/LAfuHYv2xx7ydJx66i2XLV/Dc88/z9L9fYO36DfT19dLV3UNbVw8GQyKeIB6LEo/Hw7ltbbHWA9ZudyaG16r2yIRtjCES9djW2UVnZxfV1RmMMWVXMEh4eVavXYcX9fZ4Hw6Hv93Btau1KvhXw/FmjCGbzdI/MEhf/wC+nyOVTJGpqqK+vo6pU1rZb5957LfP3syZPYsZ08M9y3HkCx1BEARBGA4iAAqCUD5oDf29MHU6nHQmXPsjEf/GiF05/j760Y/yla98hRkzZpQEiiAIKsvxt2EJPLcYtiwHJwrx6vDk2pCjhqHcEOMMvZ5z/b1oU1U4ZOU48NBmVQp8n+TW9oKwuhvlR2miiQRmkkW1YR0BLhxxz2aDSbtOEwTogvC1fMUqvvrN7/H4E3+nvqGeaDRCEAx/nQvzb2qefm4JB+w/v+TSnTG1lRlTWzn6LUeUXtvX38/ylatYuWotK1atZuWqNaxes5ZlK1bRPzhI1PPQjovrOEQiLl4kQjQSwdEai8EYO6Jr22FMWUssGmXjps1s2txOdXVGjgBXGMbAhrbNxGKxPeq78PiuQilNEIQFprL50NlnfIMf+Ph+Hq00zU2NzN9nHnvPnsmMadOYMWMqe82cQWtzk3SIIAiCIOwhIgAKglBeOA5sbYeTzoAbroR8PhQjJHDcI4FkqOPvYx/7GN/+9repq6srBHmh4y88Nlghjr9Ny+Hpu2DrqvC5IY6/nc+qKSwBGodBvums5CPm9SL+7dy2ngfdncS6O/Hd3bt9lLI4kRgYvww+wHDWBkXOzw/71WN2ZUXXX0H8u3rRr/jRT35B3vdpndIK1mJGILBZwFGKwA9YvWb9Dn8XBMVPZgpLqUMykeCA/eZzwH7zd3htT18fa9as5+lnn+PZJS+wZMlSNm/dSl9vH+1bt5IdzBGPx0nEYySSCfQorYCu69LXP8CqtWuZN3e25HWtEAJrcZRiydIXyPs+qT2oAmytpbe3l77+AfoHBolEXNJVadLpFJmWJmZNn8H8ffZm3tzZ7LvPXGoyGbyX2IeKe1X4LU94lF7JmBIEQRCE4d+bSRMIglBWKAWDA9A8BRZ8Aq67XNpkVM34Ysff5z73OS644ALq6+tLgVnZH8nb2fHX9jw8vxg2Lgsdf7EXO/52RRi+xviwWse5dg5bVFxyARbHCmBdj2jnNqK9PQRe5GVbpThmItaUSU3lYRQBAXL5iT36HQRBaW6tWbuOC/7nezzy17/RUFdLJpoeXR6+gvahHMX6tg1QEBdDAV+VRnvp5YU5bqxFF9YEgKpkkv32nct++87d4XpXrlrL8lWrWL1mLStWrmXVmrWsXL0G13VGVXTBGEMyHufZ557nXUe/VY7/Vgg2CMB1ee75pQR+gHb0iITqoeMvn8sxc8Z0Zs+czowZ05k1cxqzps9g1szpxKKRlx471hL4AUq9OC/gzuNcEARBEIThIQKgIAjlqUj0dMOxp8GvfwGD/ZILcLhNV3BDFKv6KqU455xz+PrXv04qldpBFHAcp3wD8hc5/pbBM/dA+0pQGpLV4UDZhePvpfDRuAxyibOUM4zkAhzS2OC4uIP9OIP9mGRqt4JqWHxi8l25wx2/jtb0DwwAjNrNNhLRY+i1/fKm3/HDK37G4OAAU1qaC8Pb7NHUiHpR2ts7Sp/npY5nFr8IcHa6NkqFs8PrKFZYnTN7JnNmz9zhPT5/3oXcfd+D1NZW7zZH4a7aIh6P8/iT/ywt7ULl8Pd/PoXv+zhaE/hmRAZOx3Ho7Oxm3t6z+PV1P33J8bF9Lujir9K41p6EKYIgCIIwloj1QRCEMlyZHOjthinT4T0LShqF8DJNVgjgiwn8tdacf/75bNu2jUsuuYRUKkUQBARBUBIFyhITFD9Q+PumZbD4arjvZ9CxDmJVEM+E2tMIj4UXcwGertYxTXViiOLIwArnluPgDfTj5QYxuxsb1qKiUexkjyE7PAHQGIPruAz095eEifHKRRcE24scrG9r4yNnfZ4Lv30xnuvQUF8/pMDG6CgKMJ7n0t3bRzafG+G/V6WH46jSlwCqICIGQVhl1ff9kjDzmtfMo6+vF13IpTiioWUM8XiUJcuWYZA0DpWAtbaUCmL5itXE4zGssSNSb4t7UN9AH60traWxkC+MrSAIwqrWavsYdByFM2R8CoIgCIIwDjGjNIEgCOW5Omno2AKnfrgQeBjJH7XLZtKlHH9BEOB5HhdeeCH9/f1cdNFFZDKZkuhQ9o4/CuIvQPsquO/nsPgXoQiYzEAsUcgHObpiDsVcgJDnm2oFoCUXYHGuWUO8uxOrdt8e1hq066EmOS+nsQbHdV5WLChVG3U0g4O58bsYG7ZL8Rjur2+5jRNO/RCP/O0JprS0EI1G98j1t+NUMTiOZnBwkIGBwdJze0pRjHFdF9d1CQrvOW/ObHJ5vyTYjBTXdcll8zz++N+hcNRYKF+KY2nd+jaWLV9JKpUkMGZE7s3inqTR7D0ndJQGJsArjK2i6CwIgiAIwgTf9ksTCIJQtqLEYD80tsJpHytEFdIs25snzIdUTIqeSCS46KKL6O3tZeHChUSj0R0cf2UbbO3s+Nu8Ah64Bu65MjzuG02N2vG3K4q5AM9Q62iyXYCLfpW7AG3hP052IBQAd9PM1oJXJtNRoYaVm85xNX0D/eTHIQ+gtaE7SinN5i1b+OQ5X+aCb3wbtKZxDFx/u/w8jkt//wCD/QPj2rYA01pbqa+vI5fLjeo7GNd1yeV87n3w4e3tJZTvelDon6eefY7N7VuJRqOj6jNjAlzP5YD5+4ZLvJLjvIIgCIIw6TGkNIEgCGVNbw8ctwBcD4wN87+9mhftIY4/YwzJZJIf/OAHdHd3c/755xOJRHao6ls5jr+1odvv/p/BxqUQy0AstUeOv5fCL7gAf+gsK6TCFWUZLG4+N6y7AqU1Tpkc5ywK2y/rgDMGVzvksvlC7kLwfTMmQpQxpnQNt9z2J44/9cPcu/hhWpubicfGzvW3M46jGejvp39wcNza1nHCwTB92lRmTZ9KX98AWo98PbHWEot6PPX0s1AQBEUELF+Kx3/vu/8hUlXJUfdVPh/gug5vOPC1hb1L+lwQBEEQJj2WlCYQBKF8VygHujth1hw49n2FiP9V2hRDhD9rLdXV1VxyySX09vZyzjnn4DgO+Xy+lHupYhx/7avhgWvhnitC91/R8afGxvG3K4q5AN+vNjDNSi7AYlu72RxGqeEV93iZwhMTPjd24ywKrCUajdHR2ckvFt1YOCrv7DCfRkrx32it6ezq5DNf+ipf+so3CIxPa1PjqFx/xgQlXXx3/1YphW9s6ZjuePREsX0A5uw1i/7+fkZRCJggCMhk0jzz7BKWrVxVek4obx567K+kC7ljR7NfZbM5Zu81i3gsVhhPEnIIgiAIwqTfN0sTCIJQ3quUho52eP/Hi1Hyq0oF3JXwd8UVV9DR0cHnP//5QpOEf+d5XhkLfzs5/jo3wIOL4N6rYOPzEEuPm+NvZ7bnAvS5yFkOqFd5LsCwmrL2c8Ny2CqlymoGOq4axjxSKAvfvfQK3vGe9/OLX95ELp9/0fwa3lDe7vr74x13858nn8Gd997PlNZW4rEY/ghdf9Za8vk80WiMlpYmksnksEQXrbcf0x23aVtok9cf8BqUowmC0UmNnufSNzDA/Q89Eo6hcnUmv8opjru77llMR2c3kUhkdCuKUvT09nL4wW8sjXFBEARBEMogtpQmEAShvFcpDYN90DodTvjg9ude8R97R2GipaWFn/zkJ2zbto2zzjoLpVSFOP7Mjn22dQ08dD3ceTm0LYFoEuLV4+r42xXhEdYIp6t1NL/acwGq8HbADtehU2ax/HDHvhfxaGlqoru7i29fchlv+8/3ccVV19DV07NTMZ1df8Chrr/u7l6+cN6FnHPuQgYHc7Q2N43O9eeHR/lra2tpbGwsVSgezmdSE3AL5xTm7ZuPPBzPdfF9f5TLQEBtpppb/3BH+L5SAKI8951Cv9xx973hlwJ7sNfm/SyHHPR6AIJAqq0LgiAIQlns9dIEgiCUPQYY6IUTzyj8OXjFVgTeubhHS0sL1157LRs2bODjH/94IZiuJMdfYZvpbINHrg8dfxuegVgVxKsmxPH3UvgFF+AP9XLAQ2FfnZuisaAtgeuihuFes2HJ27K49FBwsyMYloZEIsGUlhayuSyX/O9V/MeJH+TSK35Gb28vWmscR2Htjo7Aoa6/O++5j+NOPZPbbr+LluZGksn4iIUxay2+7+NFPVpbW0mn0/i+X8rtubt5ba3FYkv+v/FaBYrXMaWliVnTpzKQzY5KFDIG4okYy1as4p9PPwNyDLjsCIIApTVbOzp5+K9/o6Y6M6oclkop+vr7mTFtBq/bf7/Cc9K+giAIglAWsaY0gSAIZY/jQFcnzJkP7zqRV2JEUUy8XhQA5syZw/XXX8+GDRs488wzASrT8bdtPTx6I9x1Oax9FiIJiNdMuONvl8MKgBin6fW02g4CUhgUDgYP8+pKN2kV1ineEuz+k9sROO/G/UZG6RE576y1GGOIx2LMmD6VfD7HFT+7hrcdewoXX3Ylm9u3otSOc0xrTW9/P+cuvIizv3A+fX19TGlpHpXrzzcGi6GmpobGpka01iUBUSmFGYYwZmxAPBolFo+O/9AofL6j33ok3d3dJVfgSPFcl8AE/Prm2wr9JqpQOVEc77fffS+bNrUTj8dHdXTXcRx6unuZv+/eVKVSGGPKtxiVIAiCILzKEAFQEIQKiU40dG6FMz5ZiIAr/0iRUqrkpikKADNnzuSGG25g6dKlfOADH4CCM6MiHH92iPDXswUe/RXc/WNY81R41DcxuY6/F7U/FoMGAq5yXqBRdYO1BCTIk8SiURh04fGKliusxUQiqGEE/NYYAlRZnARWShGJxkf5kUMXXiwWo7W1GaU0V/7iOo496YN8/TuXsGlLe+m19y1+mGNP+iA33/YnmhsbSSTipQIcI/l5QRAQcV0aG5rIZDKYwLzIZTWc9/V9QyIZJxFPjHsbF516Rxx+KCYwI/7cQz9XTaaaex94hI5tnSitxQVYTgFBYe2+/qabqautGXUFa2MCLHDEIQeHf5b8f4IgCIJQNrjSBIIgVEh0An29MH0OvPtkuP13YUEJU5kBpFvIp1V0WBxwwAFccMEFnHLKKaXX5PN5XNctb/eEMWHflI76boIli2HNP0OHXywF2g1FvzIMBHUhF+Cxqo1NagNtuopbbT332Rr+zzSEQqAyhXPoPk6pfIjmFRHWGhMa/pJV9Kdrw+d2o3QqIFDlcvmGaDyK53n4vj+q46nWWmxgiUY9pk+bSt/AADf85hZ+d8ufOH3ByeRyORb96jckEymmtrbuMG9Hcp1YyGQyVKerCWxAPp/f5bX4gf+yQr/WGj8ISMTjpQqrehzzohbXnze+fn/mz5vH+o1tZNLpEYt31loSiTir1qzh17fcyic/cua4XrcwfPwgwHUc7rxnMStXraKpsXHUAmA+7xOLRTnuP98Zjh/pY0EQBEEoo9hHEAShYrAw0A8nV2YuwF05/vbbbz9uvfVW/vWvf5XEv8pw/NkdHX+97fDoTXD35bD6yYLjL11Wjr+X6RksESBCC/18Qi3j1/pJet3FPOku5jNqOVNsD1hFQJI88VeEO1DZAKJRyNSQ+edjzH70HnLxBHp3wo5SBMaUjQDqKEV1Og16z/LKlRyBkQitzc2kMil+vuhGbvjtLTQ2NFBVlRx9rj/Po7GpkapMFfkg/5LiiqtdrHnpIiBF4dH3fRKJJLFYdELWLVOovn7IIQfS29M36vEeBAH1dXXc+Jv/K723VIid5F3VWtyCyLvoxt/get6ohVlHa3p6+zj4wNeRTqVKfSyU684nCILMG+HVhjgABUGoHBwHujtg39fBW98N999eEJjKP4Dc2fF32GGHcd5553HccceVXpPP5/E8r8wdfxa0Ch8A3ZvhucWw9p8QGIimIFpVto6/l7+hUwR4GDw8LBHyHEgHB+p2LtOw0tbwJ+pYTA2/D+rxVRKLAQIgqBx3oDGgFDZVjerrZc6dNzDt749gvAh+LL7bflOA9U3Z9G/oAozT4DbQ093D4OAg7IErrnhU19WaqYU8f35BlB/pdQFUV1dTVVUVPufvxlWld195VWtNNpunrra6dL3jHZIYY9Ba88FTT+KW3/8Jf5RCq7WWRCzGho2b+Ol1N/DfZ35g2FWPhfHrW8dxWPzwYzz+5N9pbWkZtZCuHc1A/wBnLDgZgJyfJ+J60shlSlDIh2qMVGkWhGHdKypNYCym+CW4IFQgIgAKglBhaOjphjM/HQqAZXzjqpQquWeKzqGDDjqI7373uxx11FHbb8KDAK01nlfGgZKxodhaFP76tsIz98K6p8DPQSwNUR2KQrZy83o5WJxCkQuDWxJWHAyzVA+fYiufQpN34/zDVrHINnOLaaSNKIFKEmCAPAq/JMnYchIEbQBeFJIp0kueYt87b6FqSxuD6Zphi+kWi3KdsnLfGmOIeBHq6urIZgfp7ukll82CDUWJ0ZIfoeOPYpGRwOBFXGpr64jFYrs87vuilU1renp6SmLby3xajO8zpaWptH4UiwiNV9sW3/+663+NtXvm3LNAVbqKG399Cx885SQS8RhBEEihiMla8wrtfsVVV5OIJ/bIkRkEhlRVkutu+C0Hv+H1RKNREXjLjvALAxMYlAJTFDMEQdj9Pq1MeA+kFMaa0p4va5xQSYgAKAhChe2+Gnq7Yc6+8JZ3w+Lbw+fKTAj0PK9UtRfgyCOPZOHChRx99NHbxYWKcfyxXfjr3QpLHoBVT0IQQDQBkVTFOf52hypVCbalZwI8fDyiWDzyHKzaOVht5n81LKWGO2wdd9o6/mxqseXmDiy4/khVw2A/c/74a6b/7UGM6zKQqQ2Lfwy3/yxEtAqdgOU0VAtrQDQaozEaZzA7SG9fL4MDg+ENutY4Wo3rMC26aTKZDJlMBmvtsMS/wuJGNpt7WQFQKUUQgOO4zJo5Y0LaVGtNT18PZ332fB55/HFampv3SCSy1pJJJlm9bj0/veaXfO6sj4n4N4lzRmvNH27/C//41zM0Nzdjrdmj90tXVfHQY49z3Cln8tMfXczMGVOlocsIp/CFyNTWRs4+/X1EIh5WrEyCMOx7Q2PCwmmJeLwQlkhGNaGyEAFQEIRKjFogOwinfTQUAIvixiQLUEMdf8Wg/4gjjuAHP/gBBx10UOl1FeH4K7ZlUfjr74Jn7oa1/4L8IMTTEHUq3vE3osDpZdyBe9PD3morn1YOWR3lMVvDTbaBW00DbSQIVOxF7kAzQWl4lbFYLwLJJOmlz7HP7b8jvXkDg6k0OM6wKv8ODfCdSISo62LLtILrdiEwSiwWJZvN0dvbSzabJZ/3cRxnXL6t932fSCRCTU0NsVislMtzuNccjephvjbAcRVz9ppZWnfGfvrb0LuqNS8sW8mnvngeq9euY9rUqWNSudc3huaGBq795W849cT30NLcKC6KCSawFkdrcvk8P7ry56SrkjiOZU+711pLU2MDbZs3c+IHP8Ql37qQo458U+nvpI/LZD/TDo31tdIQgiAIrzJEABQEoQLvXB3o7IAD3giHHQWP3gdKT6oQtbPj75hjjmHhwoUceuihOwgEZV/VtyhYFIO0vm1hVd+VT0Dgh8U9IslXnONvJLyUO7CYOzCKz1vUJt6i2rhSO7xAhj+ZWu6mjj+buoI7MADy4+uhK+b6q0pBNsusP/+OmX97AItmoLoWZUaex88GAfHqalzPw+Ry24vAlCFFITASCY8G5/N5BgcH6OnpxvdDEX4svrkv/pxUKkVNTQ1KqREXC9FaMzg4SC6XQ+9mfQiPy7q8/oDXlP7t2C4BRZFGcf+Dj3LOeQsJTEBLY+OYiH+lNTPqke3s4qLvX8b/XnLRTj9bGG+Ko+aKn17D8jVrmN7SQhCYMZt7tdXV9Pb1c/bnz+fL55zFmQtOLThY5bh3Oa2RMt8EYXT7pDj/hEpFBEBBECqX/j44/VOhAGgmXvzblePvmGOO4ZJLLmH+/Pk7BOxa63HN0zUGdzPFDxX+PtANz94Lq/8JuQGIV73qHH8j4eXcgXPpZq5u5xwcBnWUB20tv7JNXGNbxzOyA8+DRBVVK5aw7x03U9W2hlwqg3WcUPwbIYHvE08mSadSGN8va/Fv5yAXQDsu6XSGZDJFX18vvb195PP50TsCLfhBWOG3traWWCw2YuGvNH4ch2w2Ty6bw/Veep3QWtPX38/MaVOJF44fjWUAb6xFF95v0Y2/5ZsXX0YmXUU6niIY4zQLJjA01Ndy+1/u4Y67j+aYt79NBImJmhOFfn5h6Qp+sehGmuvrx75/jSGZiBOJeHz92z9k1ep1LDzvCziOgynUURcmFxEwBGH09/+CUKmIACgIQqXeuUJ3J+x/IBzyFvjr4gnLBaiUwnGcHar6nnjiiSxcuJADDjig9LrKdPx1wvMPhI4/Pxe6/ZK1r2rH34jGxm7cgTF83qHaeIdaw1pzBHfbRjQ+YzdqDcqATVZBkGfmXf/HrEfvxSrF4Ehz/ZWGSJgkPp5MUltdjbKWiqwZaQ25XHgEOJ3OkEpV0dfXR09PD/mC8264AXGY6y+gKpOmOl1dmu+j6zHQ1uLn8qB4ydyKFnCUor9vgDe8/rVj3jxDnVlfvej73HDTzTQ1NeC57vhVCVWKmppqvvn9H/HWIw4nFouJC3Dcl/ztIu8F//MdKLhkg3E40m+txXNdprQ2s+hXv2X9hjYu/e43SSRiGBOgtTgBBUEQBGFCQ2hpAkEQKhZjIJ+DBR/f/udxDByVUmitsdaWgv3jjz+epUuXcvPNN5fEv2Lur7J3/NlCZV+lYLAP/v4HuPOH8PzD4EYgkQmPW4vjb49wsBTTrAe45IkDEY6gC/BwxkpOMwaUh62uI7FhFW+45jL2fvAu/GicXCI5olx/RYLABxSZujpq6uoKCbBNxfbF0PlrjCGVStHc3ExtXR2O65LP53f7+fwgFBEb6hupq6krCKSjb5PijVh3by+O47zkoXAFWKXoHxjg4De+rnAtwRgtBxbHccjn83zoE+dw029uYUprE57r7lHBj+H83GQiwZbN7XzpK/9TWmetfNkwjvtY+PulP/4Z/3jqGepqa8dF/BvaxwBTp0xh8UOPcdLpH2bVmrVo7VT0WiIIgiAIFXkvLE0gCELF4jjQ0Q4HHgoHHr5jdDOmAZPCLQTCxYBlwYIFLFmyhFtvvZU5c+aEwXhBFByvIgNjFI3tKPwNdMO//gx//h4sfQiUC6naQk5FCcLHdBzt4A40vEl1QsEBtsfvbQ0kkxDxmHr/nzjol1eS2byB/uoarKPRI3b9GYJ8QCyRoqGxgapkDIKAV1q4XszbmUqlaGluprG+HvclhEBjDL7vk0wkaW5uJp6I75D3c08IgoAgCF523VBKMTg4SEtLIwe+9oAxX+PWrF3H8aedyaOPP0HrlFZgYoQ4YwxNTQ3cftc93HTzrbJQjCPhMV/FE0/+kyt/fh1NDQ3YCRLhjAloaWlk3fqNnHrGx3n8yX+UHLey1QiCIAjCxCACoCAIlU9uEM48uxhljGlQvLPjb8GCBaxfv54bbriBefPmYYypTMdfbgD+8Qe441J49n5wXEhUi+NvAihmB5zHANhBAtzRb8bGgNLYTB3xtvUceO3l7POXW7GOZjCRGlWuP+P7oDSZuhrq6mrDI6D+K9OpU5zfQRDgBwHxZCjuNdY3lqp0F4/6e55HU1Mj9fX1pX+z56KIwfM8erp7duuG0lrT1dXD/HnzaGqsD0fRHuTwGiruPfL4k5z4wY+ybt3/Z+++4+Sq6j6Of865d/r23WxLoaZCqFKlS9VQBARUBBXwAUUBUXiwd2wo1kdRbM+j4POASkeaAlIUkU4CSElC2vadnZ167znPH3dmsoFAspvZ2ZLf+/XKJmzI7My5Z+695zu/c85a2ttnYCZgh+cZM5r56reuZOlzLwRrq0oqVFHGGByt6e8f4OOXfYGamgSOo6lmK/u+oamxAd/3OePcC/j9dTcUr7VI1acQQghRBbIGoBBiqo/gYWAQdtsXdt0Lnnhki9cC3Ngaf+eccw6f/exnmT17dnnwbIyZ/LsZvnaNv2wKnv8b/PsfUEhBuDao+LNGyjCqJJjk6TJbpdhBp3nR1qHwRv2ZnPJ9bE0dWJ9Zf7mZHe+/E219Mo3NKGPHUPVnMZ5HOJGgobGJiKvxPZ+tpVeoYkWg1ppoPEpNXQ3ZbJZcPkfYDRGLx/E8b8xr/W20LzgO+UKeXD6HdjZ9/D3jsdcewfp/pTVGx8K3Fqd4Tvif31/P5Vf8gEgkTEtTA55vJuA0ZQmHI+hMjo9cfBm3/N9vicUiGGNko4IKtW+pHT926efo7u2jtbVlQqbgGmOIJ+K4IZdPfelylr/6Kpdc+JHiDsEGx5HjLYQQQozb0FmaQAgx5ZkC+B68b0QV4Fim4CoFr6n4+9CHPkRvby9XXXUVs2fPLlf8lULCSTzi27Dir5CFx28L1vh75p7ge/HGoPLPyjpM1aQADw1k2ZkU4IzuYmwMOA62sZlIz1p2/58fs+CumzAhd0TV3+hiO+N5WGOoa57BjJYWwlrhe1tfJWgpJDHGkM1mcRyHmngNbihENpuveGCilCKfy5PP5dHqzXtBPp8nHApz/JKjoBgejul0adaHf1/79vf4/Fe/TV1tLbWJBN4ErslmjKGhvo61Xd2cff7Hy8dDKsO29FKwflOVT3/xGzzw8CO0tc2Y0PX3gsA3TEd7G1f94n8478JLih+oyfEWQgghxvVeV5pACDHlOSHo74O9DoKddy+NrEczCgftBIFZcVB0wQUXsHr1an7605/S1NRUnvKntZ5iwV8Gnr4Tbv4WLPtL8PpqmopTfe2ogyJRocMUdFwOZAAwbG5NmbIWYgkIR+h84E72+eX3aFz+Ipn6JowbGlPVn18oEI7HaWlro742gfV9WZx/5PRg4xcrqMbh1KUUg4ODOO6bn1Mcx2E4lWbP3Rczs6OjeNoa/YccwTlMkcvlOPfCS7n6v6+ho72VUMidFFNujTG0NjXxj0cf5+JPfbH8OiUUGrtSP/nRVb/k2j/8kZmd7eO66cdozj1KKWbO7ODuex/gne/+IF09PcXnK8dbCCGEGJf7W2kCIcS0YHwwr6kC3PTIKJgubC2uCQZES879KDaT5sorr6Sjo2PqVvz5BXjiz3DLFfDUncGAKtEITlgq/ibNxdflbao/OHybuhwXq1ptfSOh/m52+e1PWHTbdQBka+pRjH4KtzHB9N66piZaWlqIhEL4FZzeKt6c4zgMDafxPG+TYZ5SioGhId51wnFQDPJGf44Ifuaatet415kf4q5772dWZ+ekC9h8a+lob+WGW2/nK9+6svz6JQQc7SXBltvsv6+9nu/++Oe0t7ZOuqUejDF0tLfx71de4YTTPsg/H32ivFKqHHIhhBCismQNQCHEdBlNB1WA+xwM8xfDc0+98VqASgXjC7M+NPNPOgOWvJfHdtkFojGwBmMserKv8YcNiiVKAYKXg+fuhxceguwQRBLFNf6sjKYmU3fFAg67qSTYYNMNhdlo3YvyfWyiBpRDxwN3MffeW3FzebJ1jaAUepSBrrUW4/tEIhHqmluIhlx8z5Oam6q/cy1DQ8lNhn+O4zCUSrHNnFkcdtD+Y/+BCv75ryc4/5OfITWcYlZHx6SoBHujPtrR1sbVv76GaDjKJy44txwCTtod1ieZUjv93w0385VvfpcZzU04jjMpg1RjDC1NTQwlh3j/uRfw5c/9J+889mjkUAshhBCVJRWAQojpw88Hv59xfmlU8doREWhVnOprg/XvTjsHrv0r9uNfRrV3sOqJp/ju86uCnVUn9eLzpUCvVPHnwdN3wa1XwJO3B1V+G1T8SbwzOS/BHvvqfiAIAF8zKg6q/hoacZO9LL72J+x02/8BimxNbbEbjLbqz2CNIdHYSEtrKxFHS9XfBHBDIYZTKfL5/GZtcpFMDnHM4YcQj8cxvj+qauRSyHfHPffyzvd8AN8r0NzYOGnDv/Wna0VnZzs/+cVv+Nb3flz+nlQCbuLKMKJ9rr3uBj77pW/Q1NhAKORO6rYzxlBbV0ttXQ0f++Sn+cnVv96g/wohhBCiAveg0gRCiGnDCUF/L+x/KMxbBM8/W57iWxwZBTmYG4GT3gcnvx86ZkNyAFYtRymNjUf50vNruGjeTBylNtgtc5IM74pVizqoYvQ9eP4+eO5ByCaLFX8tUvE3BfgoHAxH0c/DdOCQL0eAylpsLAZumNZH7mf+PTcTSqfJ1DWilBrbDr++TygSoaGxkWgkIlV/E0RrjZfLMTiY3Kwgr5DLkUjE+I8PnBEcyzF+MLHt7Fk0NDQQCoWmRIhmrUUrRUd7K//1s18zODjEVz53KUqpLdoBeVqfU3yL4wTXq1/+5hq+fuWPaGyonzLH3BhDKBQiEokwd8ft5YAKIYQQlb4PlSYQQkwrXjCdktM/XBpRrA/DIlE4/Ty45h746GchUQNrVkA6BY6L0RodDjHQn+L7L66dbMPhERV/Oljz8Nm/wK3fhMdvBetLxd8UUwr7DlL9gA3+u1i1ausacJKD7Pz7q9jlpmtQxidbWx+sjDWmqj9INDbQ0tZGJORI1d8EcrRmIJnEGLPJ6ayu69LV2887jj6cpqaG4N+P8gOJ0rTPeXN34Owz3s267p4pN412Zmc7//unG/nwxZeV20UqAV//Pi+Ff5df8X2+8u3v09LcRDgcnjJtpbVmXVc3xx1zBG87+MBy/xVCCCFEha610gRCiOk1unagvxsOPBJmbhN8r7Ye3v8x+P198B+XQCJRDP6Gg2nAIypqLBYiIf7z2ZXlwfbEDp1sMdAbsbnHsnvhpm/Ak7eCVwgq/pyw7Oo7xQRDdYedVBrsML6vIBaDmlpmPPYQ+/7iu7Que4psbQMmFBn1Wn9Yi+95uCGX5o42murqUb6PbPA7kacnh+TQEOl0epPBhlKQzWaJR6N86Mz3AeBt4XTIj557NrNndpJKpaZWCKgU7a2t3HXPfbzrjHPo6lofYkoQSLBerdZ4vs/5n/g0P//V7+hob520a/5t/BAr8vk88XicT1xwHsj0XyGEEKLiJAAUQkw/nh+EYR/6JLz3XPjtPXD2xyHkwtpVGw3+SqwFFXbIDA7z7X+vnsAB5msq/qwNNve47Tvw2E1gCpBokoq/KSzYCCREGyl2crJQ30o4PcyC637FLn/4Da5XIFvbsL5jjioQMPjWkmhoYEZrK1HHwffy0ugTecOlNcYYBgcHN2vdP6U03b19nHziEubMnokxFneM1VBKKYwxaK24+KPn0j+YROvxDQCDndNVxYJGay3t7W0sXfYCx7/nA9z/wN83eG1bo9K1SWvFK8tXcNLpZ/Pnu//CrJkdFV0v0VFq3ANjpRRd3b2cfurJdHa0B8tvSPWfEEIIUdn7UWkCIcS04zgw2A97HQgfuABcDWtWQiYT/N0mBt8KBaEQX1y6KjhRFtcCrNKQbsOKP2Pgub/BLd+Cx26EQlYq/qYJBRRQoArsGs7BY39n/59fQedTj5Kta8APj73qzwmFaJkxg8b6ejC2GJDIJX+ib7e6u7s3a+ovxeq/5qZGPv7Rc4NvbWH+UvqZxx5zBPvtsyc9Pb3jErA4joPruuRyBQYGhvA8r3IhoDG0NDeRy+U47+OX8MOf/iJoXa2x1m5V1YC+75fb9YZbbued7z2bl19ezsyOdkwF28EYy+BQinQ2i9Z6zCH0pvpmKpVimzmdXHT+OTJAEUIIIcb1jlQIIaYba4O1/QZ6IZ1+w4q/jQ54rEVHXFKDKa4oVgFW4Qm/vuLv+Qfh1m/DYzdAflgq/qbjRdhYCMd530sPsPP1v0BlM2TrG8a81p9vLYn6elrb2oiGwxhZ629ScBxFX28v+Xx+s0I3Rzt09/TxHx84nUQstnmh4SaMrJS75ILz8XyDV6H+oZUqv67e/n5eXr4SYw377v2W4FSczlQsBPSNoTaRoL62jit++FNOP+d8Vq1eiypWqZlpPm20NC3WcRxy2RwXX/YFLv70l3Bdh8bGBny/MtWQSikKnkc6k2a3XXemrraWFa+uZl1XD5jg52utK3JctVL0DyS56CMfQhf76VRbp1IIIYSYCmQLNSHE9KXH/hmHxUI4zKeeXcnFO3aW1wKs/JDkNbv6WuDFh2DZ/ZDqhkgcEs0jAkIxrboogHE4uMVFR2Jkwi6uGUPVn+/jhsPUNTYSj0Yxnocs9TdJbrRcl+TgAOlMZvPCP8ehp7eP3XbdiQ++790wonpvy0+JQaXcrjsv5Pglx/CnG2+hs7NjTEGgKoZ+vu+TTA2TyWSwBvZ6y64cfsgBHHXYIXR0tHPHPX/lo5/4NJFICMepzOYdpjg9dJuZM3nsyad5x7vey0c/dBZnnfkedLGNfd+fVlNIfd+i9fqg9bY/3803vvdfrF6zls72VqCyU6EV0NXdzYXnncPHzjubQqHAzbffxe133s39Dz2C5xWIx+LU1CQIhUIY42PM6I9tqb/vt9eeHHvMURXt70IIIYR4zXWXU8/+gjSDEEJsZADkavyhLJFYiAOb64LKwIoNTGwQ9im1fm7f8w/Aw7+HV/4Z/F2sFpQr1X7TuY8FpX6EaqPc8WwXvak80ZDe7CNujcEA8Zo6mluaCbvutK+AmkpCIZd0Ok1vXz+Os3nVUr4xZDMZfvr9bzKjubnC5x2KawFqdttlEdf/6RY8z8N13c3srwqtNVpr8vk8a9Z1kcvl6Ghr5dhjjuLrX/oUZ5/xHnbbZWdqa2sA2GG7bYlFY9x6x1+oK36vUiyWRDyOQvHnv9zLvfc/zJzZncye2VkOO5nigZItLvUQVNvBK8tXcOnnvsoPr/o1SluamxorPvVZOw6r1qzhpOOX8JlLLgwGDI7DwvlzOfaYIznlxGNpndFKV3c3q9auo7enD8dxiEQio15b0hjD0HCaK7/5ZdpbZ5T7pxBCiI1fh2+88UYee+yxDa5zmz6xF2cX7b5v8Cs1tEWFEmLqkgBQCCHe6FqpFBbFQ4NpPj1/ZnktwC0ejFtTrPgrPs6L/4CHr4VXHg0uzrG6YMqyVPxtFawfbDzzak+aB5cP0BAL4dtNV5v6nofjhmhsbqKurhbr+7Ij6mQ5prYU/qXo7u7BcZzNCqEcx+HVV1dz7tlncNwxR2IrHP5RrgI0JOIJPN/n7nsfCPrPm/Qdx3FwHId8Ps9gcoi+vj6i4QhLjjmSiz5yDp+55CIOeuu+NDUGm9Z4noe1tjwtd4/dFjM0NMR9D/ydhrrain6kYa3FdV0aamtZtXotf7jxFl56ZQXbzJ7FjJbmcrt7nj+lgiXf90FrdLENu7t7+e6Pr+KyL1zOildX0dbWQjgcrvh73nEc1qxdy1v33ZsfXfG1chuXnpPWmkQ8zu677sxpJ5/Avm/Zg6amJtat6+Ll5SvJ5wsoBZFIBEfrN12P0HVd1qzr5oS3H8X7Tjsp6O8yIBVCiDckAaDY4j7EHx6S0YIQQrzR9VIpzFCar+09j8vmzxxzNY4tDqI2+Lcv/j3Y4GNwHYRjwS+Z6rvV8XyD2xDjb4+v5d3XPMmOLXE8/437gDUGYwyx2lrqGxpwi1U0YvJw3aDyr7u7G0dr1GbcZGut6entY/HCBVz7658Ex7oYolXayOUMjjz+NLp7+qitTWyw9lqp2s/3fQaTSYaGhmlubmTxTgs59cTjOGj//YjHo+XHNMZiVTCtfeRzHvkaPv3lb3Dt//6RWbM6x6XPOkrhWUtXdw+RSIglRx3JR845k1kzO8v/TynEYhJWBlobbNgzcuryYDLJT3/+G66/+Tb6+vuZ0dxCOBwqrwVY0fZzNGvXdbHr4p35/a823gdLg01jgrUt13/f8PA//sVv//ePPPHU06zp6iYcCtHc1IjjOFhjytfBUtCXy+XwjeH2639LS0vztJu2LYQQlb5GKKU466yz+MUvflFehmMzT/Dg+8HmiB+4ANauAldWg9sq71GlCYQQ4s0GyhYiYT6zdBWXFasArd38HTkt4FlLqFjFAcDL/4Rl98HA2iD0q2ksJTvS4FshpYC8Ye6MOJ11EbIFQ0irjVZJlXb4bWxpIRGLgaz1N+k4jsPg4CADAwObXfmnlCKTyRCLxfj+FV/Z4EZ/XPrciKnAn7jgPD580WXU1ibKGzv4vs/wcIbhzDDGWPbcdTFHHnYIhx96ILNndW5whvM8H8dx33Dqp1Kq/Fq++tlLyQyn+dPNtzN79syKh4C+tShgZkc72WyOP950Kzfe8mcOP/RATj/lRPZ6y+4bBEylHYonMnSyxTU8S32l9FyWPvcCf7r5Nq6/8VaSySEa6huY1dmJ53njFP45rF3XxYL5c/nVT66EEX3ktcezNJYE8HwfXQyL99vnLey3z1vIZrPccsdd3HrHPTz48D/xjSEei5JIJAiHQuXj3ts3wMc/+iFaWpqx1kj4J4QQQowzCQCFEOJNB2fB9EwznOZLy17lcwtmYbA4m5igaYv/VisIKUUaGHr5KdqW/RkGuiAchZomqfgTwQC7UKCtJcaOTXEeX5tkRjyCP6JfGGOw1hCrqaGuoZGQUrLD7yRijCHkuqAUPT09ZIobfmxu+Of7Pn39/Vz1/W/T2tIyruHfyJ8LcORhB3PgfvvwyGOP0VBfz6o1a9BKM2f2LE5YciSnnvxO5u2w3Qb/tlRFp5TarPUDVXH5BEcpvvP1L+KGQ/zfH29izszON50iOlae5xEKubS3teL5Prff+Rfuvvd+5u64Pe9ccgwnvOMY6upqNnju5Sm3r2mfyl5P1lfPgSn3kdLz8DyPu/76N669/gb+9fiTDKfTtLY009nehrG2Yrs2v1Yw7beLxYvm8dur/4tIJLzZa/G5xdCuVL2olCYajXLScUs46bglrO3q5qZb7+DGW/7M8ldXMZRK0drSQsH32Hab2Xz47PcXG1ymogkhhBDjTaYACyHEpgIaFKbgEY6GSC3Zm5CmPJh93QBvRMVfyU9f7uIjz6zhp8l7OCuyDi/cgCvBnxgZWPgGtyXG169bxtWPvMqshmh5GnBprb/apkZqpOpvcp4jtMYaQ3dPD7lcbrM31Sh5dfUaLr3wI5x71hnBeaQKASAjdspd9sILHPr2dzFrZidHve0QDj5gH4449OAN+2gFquU2mA78xW9wzXXXM7Ozs1wlOF5c16FQ8BhKpUins8TjMQ7cbx+OOOxA9tt3L1qbmzf6XD3fR5XDQI3WNrgibOLQlKZSB8G9xdrgOWzsmKZSKR74+yP8/ZHHufMv97FmzRpi8Ri1NTVEIhH8cV7bU2vNmjXr2GvP3bj6x98hGols8UYcparG174P/vXEk9xx973c+Zd7efSxp7jml//FMUccJht/CCHEKK6hMgVYbNE9kTSBEEJsYjCHRYdd8oPDfP35VXx2wczX1f+9tuIP4BfLu/jPZ1bS3TcEoQShmkYgWQz+JPwT6ykF5HwO3r6R//7XKjzfBmv9WUu0pob6hgYiSo1bBdCEvObir6n6bghCC5dwWDM4kGQwmcQYM6rwTyvFqjVrOfPdp4wI/6q3Nl0pzFswdy43/O4XzJs3l4b62vWv0VqCyItRh5ob7+cjpgN//lJqamL8/Ne/o7V1BiHXHbegq7QBSEN9HY0NDRQKBe669z5uu/Me2ma0MGt2Bwftvy8H7LsPC+btQDQaRSkVVHW+rteur3Z7/QHVOMXpsCPbd6R8Ps+Ly1fw4EOPcN8DD/PSy8tZ19WNMYbmliZmzuwMpmiPY8XfyMHkqtWrOeyQg7jqe98sH/MtDeNKVY3WWorFjjiOYo9dd2GPXXfh4+f/B489+TT7vGWPYrNJ+CeEEEJUgwSAQgixmQMlYmE+t3Qln12wfi1A1Mg1/oL/92cvd/GlpSt5tW8IQi5OTQwfl17rgPKAsDSo2ICjNeQN+21bTzTkkssXCEfCNNTVUVNTA8bDM9MjNNZaY7WmYC14HmgXx1G4xWopM8krY0vVSpFIhFwuR1dXklQqheu6o6qO047DqtWredshB/KFT11cPs9M1MYUe++1R/nPnucF6wGOw3NRSpU3Ibns4o/R2dbO5Vf+gLqaWqLRyLiFgKVKPAjCwI62VqyFTCbL0uf+zdNPL+M7P7yKhro6Fi6Yyw7bb8P2223LnM6ZtLfNoL11Bo0N9aji1OdNHeuBwUHWdfWwrruLVau6+PdLL/HS8hUsXfYcXd19hEMhQiGXWCxKZ2d7eSp4tTb0Mcawbl0Xp5/2rnL/q3QlnlIKB6DYVL7vo5QiHA6Xwz8hhBBCVI8EgEIIsTmDR0C5LnY4w2efXcmXF80uB3+lir9frujmU0+vYG0x+NM1pV19fbAw6LjlxxLidTwD8RB7zarln2uytLW14ujps9afdhyMtSSHh8kMDwdrHBoDSqOVIhKPUZOoIey62Ek8zTkUCmGNZXBwkGQyie/7hEKhUT2GUoo1q9dy0H778tNi5dVEhn8Avm/Rms1e129LqBGv98zTT2HbbWdz0WWfw9rqfTjiecG0qUgkTCwWKZ7kFfl8nqeffY6///Nx8rkcsViMeDxGIh4jEgkTDoUIR8KE3RCRaBTX1XieTyFfIJfPkc975At5crk8mXSGVCZDOp0h7LrE4zHi8TizZ3egijWwxphx2dRjU/0vOZTiPz/xMc4+4z3l/jfelXjOa9YLlE0/hBBCiOqSAFAIIUYzaA27XP7CGr64aDah4q6Xv1rRzZeXruSlniS4LqomikKVK5mcYMRFnw2BkvhPvAlr2aajhRfyWcLKUPCnx8tyXJdUNktyoB8/X0ApxfqoweADw4N5MqkU8ZoaampqCDkO1vcnRUWgKu7irZQilUqRSg2RzxdwHGfUYZnWDqvXruWgA/bh5z/4dtACk2ANNMdRVW/T0hqEu+2yU1AZOgEhaLBe3fo+5roOjQ31NDc1ooq73HqeR97zyGaz+MZgfIvFbLiOq1IoNFortKNxtEM4GqEtEccp9hFjTPDLn7iJ76pYhVvI59lnz6AKz/O8cQ99X3vsJfwTQgghqk8CQCGE2EzGWpxQCH8ow+XLXmX3xgQffPQl1vWnwHHQiaDiz1iwGxnc9RMq1RJKHaB4fRBhgp7RkQiTzw9hIqFyb5mqtNYYoHdggPTgEEpvfF20UvRlrSU1MEAmlSKaSBCvqSFSqhry/aq+a4wxhEIuSml832d4eJjU0BA5z8MZQ5VcaVOIV1et5rCDD+BnP/hW+TVvrWuglcK+q3/9OwYHh6jprKl6Ndzr3oc2mKpaWn+vFFY5joOKRIL345uFlMV1E8s7/lqLn89v8Hon9PWV2l4rfvnba/nO175Q1fBPCCGEEBNHrvhCCDEKfnEtwM88vQIKHiiNSkQ2qPjbWLjhoxjACQJCaUaxsb4FuAr2nl3DVU/2B5tBTOHXo12XbDbLYH8/+Xwe13HY1PapSikc18Uaw3AySTqZIpJIEA27RBM1hJTC2uAdVMm10koVeKUqP601WmuGhoYpFHIkk0NAUJ0WGkPlUqnabfXaLs447SS++OlPbvBzt0bW99HFtrzvoUeIRWOT6vmNDOtKYd5Y1yecDMHfBuca36e+ro57//YQ+UKB8CinsAshhBBiapIAUAghRp1sKBQK5QbrVb1Rxd9rIgZSOIDCkeo/sREOQN4wvynEgsYwa4YK1EU0/pTqLhatHQzQNzhIOplEjWEHWaV1cd8ASy6VJKsUTipFxA3WX4vFImgdPKbjOOUwMNjo4c0bzBhbfJ66HPqFQqFgTTgb7Lzc29uL7/sUCoXyWmVKjS2o01ozPDzMYDLF5y+5kPe/77Tyc92adz/1rcUFHnviaZ5+dikzO9onvPpvaxIOh+kfSPKnm27jlBOPw/P9IKQXQgghxLQlAaAQQowp5thw+afNMYwGnGId4NSu7hKVpzSYgsWpC7F9c5QX+/PURabWa3Bch0yuwEB/P14uF0z33aLqJ4UuhofW88h4HulMmmRSE3IctOsSiUQIuSGUVuhiBeGbBWvhsMLVLrl8nmwmg+cbtKNJZ9Jk0pn1P7lYCeg4mrEUfgUvW9PT00ciEeeq732TQw9+a/BaJnjDj8nRV4LjevPtdwRhrFJSHV3N9nccHK35891/5ZQTj5M1+YQQQoitgASAQggxzooTFhnGARzUZtQLiq2TBfB89p0V5+4XBzFTpKNoNMZR9A0mSSeTMCLgqRSl9Qahed7zwPPIZNIoVZq+C0qVKvuc8q62wa6jYG1QYVZaj8/zPKyx+NbHddyNhiBjCf+04+Dl86xZt5b9996Tb33t83S0tRYfT8K/kW1w25330FBfj2+k+q+afN+noaGORx97khUrVzFn9szypixCCCGEmJ60NIEQQlRHGrf4uYvUuYiNc3QwDfjQ2TGiIZf8FJj/q1yXnF+ga103qYEBdLFybtxvYIpTeB3HLU/lBVWeupvPZ8lms2QyGbLZLPl8lkKhQKFQIJfL4Xle8O9dh0g4UpHnrJTC0Zq+3j76Bwe58MNn8z9X/4iOttZg/dBJuB7cRChN9b31jnvo7RskGglj5VORqrLWEolEGEgmueue+0tvKmkYIYQQYhqTCkAhhBjvgVbxawaFRaGKG4FInYXYKANOzGXPzjh/Xz5EU9yZlOsAaq2xwGByiOGBfhjDzrjj9bxG/R7dwvSptHlIOp2lr7+X3Xfdhc9f+nEW77wwOKTG4mgJ/so3n8V+ctsdd4GyOI4O1mAU1T3VGENzYyPX3XQLHzzzNBwJp0Ul7nmKa7HKhx1iqvdjLR+KiOl4DyZNIIQQVbmVII0mgyYuFYDiTXgGXG05crsEd76YpHkSPkfHdckWCiT7+8llMmjHGf1gr7zmm0bbqRn+KKVwHIdMJkNP3wAtTQ187lOf5MzTTi4PIAC0hH/r+3dxs4nu3l7+8a/HaaxvwMrmHxPCGEMsFuXl5St4+tll7LxogUwDFltscHCQoaEhHMfZ4g9XhJjI63tnZ6c0hJh2JAAUQohxp4sBoEsahzj5Yv2f3BiLjdx0AhRg19YoM2td0gVDJKQnxRRJrTVWKQaTSVLJJNaYMa31Z5TCLeQJ5/Io61OIJfBcN1iXZAoMGEsbhORyOdasXUdtXS1nve9UPnruWdQkEsFrNEaqB96ofwN/ve8B1nV1s9022+B53rgcIypQ3TlZ+tt4vZZQyMXzPK6/4RZ2XrRAqrZERfrryF9CTDUSXIvpTAJAIYSozu0EvtGklQNKNgERb8zRYPKG+sYQO7XGuO+VIdpDmomukXLcYOfc/oEBCuk02nXRo60UUgprFdHUIF44ytKjjicylGSbR/5GbDiFH4vihaMYrdGTsCrMdRyMtQwPpxkcSpGoifHeU07k7DPfy6yZHVAM/lSV1kGckv272Gf+70+30ljfgDGVr4hWSpHP5/F9n3g8jtYa3/em3DqDTrGyNpPJlNfsq/TA1FqorUnw4D8ewfM8XNeVKkCxxe8/IaZ6H5YQUExXEgAKIcQ4K60BiHJJF6sBhdjUoBwsR2xfywPLUxO6BmC56m9oiNTgYFD1FwqN+nGMUrhegfBwit7t5rL0qJPIbrMjeB4r99ifzif/yawn/0G8rwcU5BK1QRC4YaNUqf3t66pYfN9jXXcP+UKe2R0dnHLycZz13tNobm4q/xvGuAbh1tOvg3Z9ZfmrPLN0GfV1dRUPAJVSJIdSNDbUEw6FePHl5WjHoaWpgXA4DNZiisdqMqxVZq1FFXe5ZkT1VMHz6O/pJZUeZmZHB4lEgnVdXdTV1lb05/u+T31dHc+98BIPPvwIBx2wn2zIIoQQQkxTEgAKIUQVKAwWhxQhCQDFJjkayBoO2iZBQyJMPu8RdqsfUjiuS87zSPb2k80Mj63qDwVKEU0PYZTDssOP59W3vg2wqJ51oDV+bR0rD1vCyn0PpvXpf9H2/DM0rXiBSDaD70YwIRcTCgfBzThOrS1V7mmtMcaQyWTIZHJkshlCoRD777MXS44+nBPecRSMCI2kYmrz+NbiKsX1N95MoVAgHA6VdwSu1PFLpVIkEjFuuPaX1NXWcuOtd3D3vfdz/4P/YO3aLuLxGOFImFg0SjgcRimFMT6+X/21WbVWOE4Iay15r0AukyOXy5HOZolGIuy31x4ccMB+vOOIQ5nR0sLxp57Jv19eTuuMlopOm7bWEo/H+MNNt3HQAfvhutKXhRBCiOlIAkAhhBhnFggBBRySxQpAiQDFJvuNByqhePv2Nfz6iR5mhkJ4Veo4Wmus1iOq/vwxV/05foHIUIrebXdg6dEnk529HQwlwfewpdAsmw1+hUJ07XMwXbvvh9vfQ+u/n6XjmceIDvYSGU6h43G8aBSTL+CZ9cGR3uAPbxwOliq9yr8H/1H++3w+T2p4mNTQME7IobmxicU7L+Dotx3C0cUQpvzaSpVbxY1AxCb6s7W4xeD2vgcfJh6LVvTxlVLk8nny+QJX/+g75Uq5495+JMe9/UiSw8M8+cQz3HTbn/nnY0/RNzDA6jVrcVyH2tpaErEYrutirR2XacnlLqoUSms8zyOVyjI8PEwml6Ohvo6G+lr22mNXDj/0IA5+6760tGy4BdDPfnQFx5/yfgYHk9TWJjCmMicE3/epq6nhvgcfJpPNEotGZRdXIYQQYhqSAFAIIaqgFDcMyuYfYnMH5YBr4Ijt4lzzjINnLGg17t1HO5pCwaM/mSSfSqFdB+2M/nbBak10eAijNc+/bQkr3nokaIUa7MMqDSOr+Ep/9n1Ufy/WcfAamli9zyGs3v8wGOijZfWrNP7+Kmqz60i0zSIRT2CMj7UGU5zKiTFYwBpbDtrViJDD932MMfjG4HsGz3jB736BfKFAS1Mzuy/eifnz5rF45/m8de+30NTUtMHr8jwPx3HQEo6MijEGx3F49PGneGbp88xsb69o9Z9vDN09vVz59S+x+66Ly1Oyfd/HdV3qEgkO2H9vDth/bwCWLnuBJ555liefWsrSF57nxZdeYSg1TE08QX1tDcrR49IGyXSa1FCKUCjCDtvNZsG8/dhlp/nsstMidtttF17bqzzPQzkOGktrSws//M7lnHnOR8mEQ0TD4YqtUxUOh+kbGOQPN9zKe089sdxuQoz63F86HwsxhfuwENOVXNmFEKIaNxPF33sIAQqJDsQmL9AabNZnm/YY+85K8NCKIWbE3XHbDERrDUoxlMkw1N+H9ca21h9KoXyfyNAAAzO3Zenb30V69vaQTkGhgN3E9N1yVWAmjQKsshCO0LP/IfR8/4vw4jKc8BNEwxHqm5pI1NQTrakhFq8l5DjBWmpao5TGUQqsJVfIoZVDKBRGKQiFQkQjEWZ1drDjDtuy3bbbsGj+DmwzezaJ4i6+I/l+aX0/JBQZo1KV5E23/hlHa5RWVKozK6VYs2YtHzv3bI57+5Hl7wHrq/oATHAMlVIsXDCXhQvmctpJxwMwkEzS3dPLn266nV/+z7W0zmip+PTkdDrN2w45gA++593M7GyjoaEOpfTrBp6lAsQN+1vQl/fcbTFf+uwlXPK5L9PW1oqu0NVEa43rOtz1l/t576knlttNqgDFmK8nQkxRct4T03p8IU0ghBDVYooBoBCb2WMsOBZOWVjPg8uH8M2IkrYKD9byxpLs7yObSQfr4I1hHTCrNKFMCgX8+8CjWX7QUeC6qMH+IPgbzaBQ6/Ub6OTzqOQQ2CBM9/NZhvNZhlODr/9nroPruDhuCEe7+NYnkxri57/8Nae86yQ0mkQi9ubtbky5Yi2Y4iuDgUr589330tBQX7FwzXE0q1ev49hjjuDCD58d9JjXBFdKKRyAYpcuTfM11uIWj3FDXR0NdXWc8e6T+e211wWVdxUeBBYKPu856Z3svNP8Db5f+lla62J/e8NRKQAnHvd2li9fwQ9+9mtmd7aXNzXZEr7v09DQwKOPPcHLK15luzmzpApQjEldXR01NTWyk6qY0iQEFNOVXNWFEKIKTDHI6LbhoEJKpgGLzQk3NJDz2XVWjAXtcV7sylAfcyq2S6cuhnKpdJpk/wDW98a2nl2p6i81wFD7LJ455mSGt18Iw0nIZjZZ9bc5j29dZ3346YZQvrfhu6jYKMbzyXs+5HIbPMTszlZqR1T3lQamZsSXUtVKaSMQURmlTVJuvv0u+vr7aWudUZFgwHE03d19LFowj+9948vl47qpgVtp3UZnRF8ohV01NTW0tc+gt7efmppExTYH8X0fpWHbObNhxFRyGF1Vaen1XfTRc3ll5WpuueMuZnV2VCRQjbgua9Jp7rrnr5zz/tNlbUsx5uBE+o4QQkxOcncrhBBVY+krVgDKyVdsfnAQJIH/sWsjGVMsC6zEDYDWFIylp7eXwZ4eMP4YdvgN1vpz02lCmTQv73cY//jgRQxvuwMke8EURlf19+Zvn/WMCQKkkb82HIGiRvwCyGbzABQKhXKIopTCKVb4lar95FP/cbjZLPaB2++8uxgO6Io85kAySUNjPb/8yZXFbmHGdPyUUjjFEC4Rj9HWOoNMJofWqmJhu+d51NbU0NISrCnpuu6Y+tvIqqrvffNL7LrzIrq7eysSWHu+T1NjPX+48ZbyzxJCCCHENLonkyYQQojxV5rK2F0MAGVYJTaXo4G0x27bJjhwTi29WZ8tmZGqAe26pHI5etatJZtKoUtr542GUmAMscE+0s3N/PP0c3npHe8CC2poCJSeuNuM4iL0IxejH7kDsAQb1eP7Pkopenr7+ce/Hqehvn6Lq+qUUmRyOXzP57++ezmNDfWYLVx3TBVDOoBZM2eSL+SK/XfLE0CtNdlcjnk7bF+RNlVKlXcq/tF3L6epsYHBoaGKhICxaJQVK1fz5NNLy8dPCCGEENODBIBCCFFFpU1AhBgNawHf8pG3NKIcB88bWyihtcbTmt7eXgZ6erDGlCufRvV8lMLNZghnh3lln0P5x/svIrn9QhgcgEJuy6f8imnnjzfdyoqVq6mtrdni6b++b+jv7edrn/8Uu+y0CGttRW9o5++wPdpxx1xR+FpKKbLZHAsXzK3cDbzW+NbSPmMGP/zO1yh4BbLZ3BY/32g0Qiab5+r//t3697t0XyGEEGJakDt0IYSo2unW0EtYAkAx+gBBA3nDzLYIJ86voyvrjaoKUAOO6zKczdK9bh2ZVKq4E+sYqv6A2GA/2Zo6/nXK2fx7yanBE0wNBH8v4Z8Y2feK/SEWjdLe2sryFSux1ox5jTCtNWvWreX8c8/m2GOOqOgmA6XwbP7cHXAdpyLVb9ZatNZkMlnmzwsCwIptgKIUxlh23XkR3/jCZ+jp6ytXXI7ltSulWLt2HY6r6exoD36G48gVSwghhJhGI1IhhBDjbMMpwDKcEmPoQwbI+py7RxMzG6IMbmYIWKr66+nrY6C7G+t5Y6v60xo3myaSSrJ8zwN4+KxPMDB/Fxjsh0IOlCz6Ll6vFEadftpJ3HXj7zn1pOPo60vS09ODUnpUQaDjOKxavZZj334EF5z3wdf9jC2+KS6GlQsX7IjjOOVptlv6+o0xOK7DDtsEG4BUNrQMfl9y9OFcfP65rFmzbtTPT2vN0FCKVWvWsN/eb+GGa37FpRd+RDqvEEIIMc1IACiEEFVjSZpQMQ40GGkQMZqBugbjgZPQXLJXM2lf4b3JhiDlqr9cnu61a8lsyVp/1hBNDpCrqefR087mhRPeCyEHJVV/YhRaWpr48mcu5Zpf/phdFu/E6rVrGUwmg2rUTYR4juOwbl03u++yM9/7+ldgxPqCFXuPFR+rvq6ehrpaCgVvix9fKUUmm6OzrY22thkb/JxKPedSoPjhc87khOOOYc3ars0KVrXW5HI5Xn11Ne3tM/jJ97/Fz354BYsqOFVZCCGEEJOHK00ghBDVYgGn/NmL1AGK0dIK7LBhj7k1nLi6ieue7mVmnctr91TQWuMByb4+0sPDqGKAMlpGaUK5LG4uzcrd9+WFw0/AJmohOQjWYLVU/YnN51uLoxS7Ll7E767+MTfffgdX/vhqlr+ygqbmRuLxKJ73+o9GtFYkk0na2lr4+Q+vCPqmtWOeRrw55s/dkb//8zFisegWVexprcllc8zabg6tM2aUv1dJpRBQKcV3vvoFVq1ew1PPLKN1RstGpxs7jkOhUKCrq4e6ulr+8+KPcM7737f+OPn+uLatmJ6MMWiteeGFF3jllVcIh8PSKGJM57NMJsPChQuZM2eONIgQFSYBoBBCVIEtf1UMEKWBjBRhizHeHQMZnwv3a+T53jTLurLMiDv4xYzCcR2GszmSA/14+QKO1uvnCW7+HThYSzzZT7qukaePfTe9i/eEXKa41p9TXJhQiM3nFPuh71scR7Hk6CN5x5FHcOVPfs5/X3Mdq9d0MWNGC65S+NZiAddxyGSyeL7PD759OfX1teWgYVzO1cUgbeH8HfnrAw/RpBq26LG01mSyGdraW2FESDIeg+bSY1/1vW9y7Cnvp39gkIb6uvJU5tI6fz09fRhrOeG4Y7j0wvNpaqwvPzellIR/Ysx9ECCVStHd3U08Hq/INHqxdQmWJBhi++23H9dzphBbKwkAhRCiqixrbYgGlcEU6wGFGNUgC7AeqJDiywe1c84tKxnKGxqiIbLWMjAwSGZoaIuq/sLZDE4hy6uL9+L5I47DNDTD0CAYA1L1J7aQ45SCQIPjaC768Dm86/glfPeHP+XG2+8gHo3S0NCAAgoFj56+Pn50xVdZvGj+uA8Gfd/HdV0Wzp9LJpMJdtvdok07LMYatttmzrgPZrXWGGOor6/nqh98m1Pffw6ZTIZ4PA7WMpQaZnAwyZ577MJlF3+M3RbvVH7NWmsZZIvKDC5dl1AoRDgclgBQjOk8Fg6Hy+ejSi6ZIISQAFAIIapLKboIswBTinKkTcTou5EGkzU0N4X44qHtfPKudawYSOPmh8jn8ujNWFNtY30Ta4klB8jU1/P8klPp2W0fyGVRA31Yx5G1/kRFOY4uT6+dNbODKy7/Au8+9UQu/84P+ddjT9DcWM9Af5JPXHAeRx9+WHlwOM7vLgDmzJlFLBrB87wtON0rfM/HdUIsmLd9VQazWgdtumDeDnzjS5/hgks+izEwkBxg9sxOPnPJhbzz2GOgdPUZ56nUYutjjMH3ffL5vDSGGNt50/fL4XGpKlsIURkSAAohRHVvbVhDGLAS/YktG+grsGmfxTPj/OwdnVx++zJufz7F7IYYZpSdyyhNKJ/FTQ+zdvEeLD36JEx9U7nqz0pAIMZxsMeIdefestsuXP+bq7juTzfz5W9dyTFHvY3zzj6zagNBVZza3jajlfa2NnK5HJFIZMzrABprcV2HBTvOq3qbHn34YZx31ot8/yc/52Pnns15HzqTSChYl61ciSgDa1EhpffnvHnzmDVrlgTLYswKhQK1tbXBvY588ChERUkAKIQQVaIBg2IdYZA9gEVFBvpgMx6zZ0T58ft245P/+wzXPdXFnPooZrMi5qAKNZrsJ1dXz7PHnEz3rnuDV0AN9EvVn6iaUlhQCgJPPmEJbzvkgPIgsFpVIKXu3tLcSHtbK8+98CKxWGzM04A9z8N1XebuuF1VB7Ol9rrwI+dw6snH09G2fg1CpZQMqsU4vHeCPhWLxYjFYtIgQggxCUkAKIQQVROEfqUA0MoUYFEBSin8YQ8nrPjWqYsIOYprH1/LzPrIm+41bZTCzecIZdOsXbgbzx1xAl5LG6SS4PtS9ScmhOM4WGsxxtLY0LBBP6/W+6lUHdfZ3saTTz0T7KrteaN+Dlpr8vk8c2Z2TsjrKOloa5XgT1T3bqfY34QYi9IGSkKIypMAUAghqkQVv66x0TfLZYQYNUeDzfsoNF87aSfCjuY3j65mZn10oz3RYommkhSicZ467j107XFAsepP1voTk+BcqRSOo7DWTshAsBQAbr/tHKwae5ihlCKTy7Fg4bwJa0trwCgb7AYuRJVIeCO29BoghBgfEgAKIUS1bogBH0UXbrAzpFT/iUreMGtdDAENXzgpqAT8xSOr6KyLoFDY4lp/biFHOJ2ie+7OLD36JPKtnZAalKo/MSkHgRMxECz9zPnz5uI6IYwZ/fTfUnCZyeRYOHfH8veq/XqUBkc+cRJCCCGEBIBCCFFthn5cwJETsBiXwb7N+yir+PTxC4m6Dj96aEU5BIymBvEicZ59+2ms3vutYCxqoFeq/oQYoVS9tHDujrha4ftmDBVNQQWj5xWYN3cHKK5v6Lpy5hdCCCHExJC7ECGEqBJT/JrEARyc4k7AUpshKklpjS0YlC1w8ZJ5hFzFD//6CrNUlt75O7HsyBPJdcwKdviVqj8hXv8eKlbpbbPNLMKRCL4ZfQCodbCTZXNTIzPb20oPLI0rhBBCiAkjAaAQQlTZICEgVIwEpepKVJ7SYD0flYaPHTOPuDGcE94VDj0MCgUY6AsSCqn6E+JNbTdnNi+uWEE0EsGYzd+93XE06cww7W1ttLXNKL4v5f0mhBBCiIkjAaAQQlRJsOKfIWkcssohqvIYNFJ/JcaD0hqMheE8Zx+1Ha/YOXx12Ac/HVQJShMJsUlz5+7Av556hlgkgud5m1fFZy2u6zIwOMiCefOIRiIAaKkAFFuBFStWsGbNGsLhsDSGmJR83ycej7No0SJpDLHVkQBQCCGqRgOWpAoxQIh28lhUORoUYjxYY1HDPl/hYeoZ5BK9GEseJdvQCPGmA0THcVg0f0cUFtd1sJv9hgkCQK00282eCYDnebL+n5jWSrtnd3d389xzzxGPx0dVNStEtXieR2NjowSAYqskdyJCCFEltvjV4tCPQztyYyzGnwLQDlDDJ3mekIGL7GIsOVRxHUohxIZKa/69+10ncsRhBxOJRoKK2s1+AIWXz1OTqAHAkbU2xXS/1hQrXKPRKDU1NSQSCQkAxaTkeR7xeFwaQmyVJAAUQogqsYDGxxChn1AxDBSiev0PYlyon8ezik+anSUEFOINlMIMx9G0tc6o2OMJMd1ls1mGh4cxxmCtXF3E5CMV2WJrJj1fCCGqyAEMDt3lTUCEqI718UOMT6hlFBR8yu6EJS/T0IV4A8aYLQrvrLWj3kFYiKnIWotSim233ZbGxkZCoZBUAIrJdy+kFL7vEymuzSrE1kYCQCGEqP5tMqtsGJRGakJE9XsfQIzL9DIGbIhvmvlAZoOIUAgR2NLwTir/xNb2Xmlubqa5uVkaRAghJiEJAIUQouoMK4lBcQsQIapJlXtejG+oZ7jfNvCQakWRwSKVSkIIIbbgDmcLq2aFqAapzhZbKwkAhRCimjfGwW0HrxIGNFqmXooJoAAPjYvDtc7TbOMfgFWh4hqVQgghxNhIqCKmxH2QhNRiaz1HSxMIIUT1BHGf4VWiUgEoJpSLxRBmjhrg187TQFjCPyGEEEIIIaYpCQCFEKLqp13LKiJyChaToDdaIMoZajlvVyuBiFSlCiGEEEIIMQ3JFGAhhKgiW/y6kojsvComBR+Ng8Mf9ZPUe41kVQRNHiMB9cSdJ6xlYGBAGmIK01pTV1cn08zEVnn+kt1/J/78I+ceIcTGSAAohBBVZ8iaMCgHlMGgJWoRE8bB4hMmTJprnKW80+wp4d8kGECnUilpiCl8/EKhEHV1ddIYYqujlMJxHGkIIYSYhCQAFEKIag4My3/SrFNR2sjKOoBiwpWmAp+gVnKo7eQvqhOHNL4EgRM6gJYKjqnJGCMVOGKr1d3dTV9fH+FwWCoBq30t15p8Pk8ikWDWrFnSIEKI15EAUAghqi6IAV+2cdpUGoOLI9OBxQRavyuw4mfO8+xoZuDjoLDSMyfqLGGtBEhCiCmjFHy/8sorPPXUU8TjcQkAq0xrTSaToaOjQwJAIcRGSQAohBATQWleJsK+GAlYxCS5IbBYIuygejlVreH3djaajFQBCiGE2PRtTfEDi0QiQVNTE4lEQgLAKisFgLW1tdIYQog3uN8XQghR3Rs0DAbLS8RBAkAxiRgsDi5fUy/xe9sp4Z8QQohRSafTDAwMkM/nJQCs9v1lMQCMRCLSGEKIjZIAUAghqswBDIoXiUoAKCZd3wSX7VU/i20/T6lGNB4yhBNCCPFmSssWdHZ24rqurAE4AbTWFAoFEomENIYQYqMkABRCiAmy3MZAIZuAiEklWAvQ8h96DefbNjR52RVYCCHEm9I6uE50dnbS2dkpDSKEEJOQBIBCCFFlpc/Du3GBEKHiRgsSBIpJNJRjierlfJPFUy5KKlWFEEJszj2OMbKB0QSz1pYDWSGEGEkCQCGEmBA+PYSBMGAw6OL0SyEmVrAjdZhtVD+760Eesy2yGYgQQojNIsHTxJMAVgjxhudoaQIhhKguW/zaZ1x6rVusCZSbNTFJBg5AAQX47EIKUHKzIIQQQgghxBQn9/RCCFFlFg345FSErmIFoEyvFJOJKt4i7KuGAIMntwtCCCGEEEJMaXJHL4QQVWYBBwOEWUtEAkAxaW8R2siV+6wQQgghhBBiKt/dCyGEmKCTr+IFGwWsTAAWk0qpPybwpTGEEEIIIYSYBmQTECGEmDAez5IAtASAYlJKyW3ChFFKyULuU5RsgiCEEEKIyUju7IUQYgIEdVWWpdQCurjzqhCTQ9AbDS8QhWLFqpFmqV77W4vneRIATlHGGEKhENZaOYZCCCGEmDQkABRCiAkRRCpLiUlTiEkniCx8HqWuGFAbjKwaUr32V4r6+nqpJJuijDE4jiPhnxBCCCEmFQkAhRBiApQqrFaaeHFBwCBgkeG+mGg+4OCDTfBXvwGUJysBVlkpABRCCCGEEKJSJAAUQogJUAoAIUQ3CWaQkXUAxSTpmwoocC/tdKsaFDmp/psAxhipIJuq7yFrpXpTCCGEEJOOBIBCCDHBltoEM1QKgytrAYpJc2PwWbsdABqLL/F01UmANHVJcCuEEEKISXl/KU0ghBATNEgsfllGHLAS/YkJFwR9WR60M7jftAN5fLlVEEIIIYQQYsqTu3ohhJiwE7ABNM/aBOBJACgmVLD2nwe4nGIXg/JwZO9fIYQQQgghpsn4UwghxISegJ8jXt53VYiJEkw/z3O22YlVtgEoSPWfEEIIIYQQ02z8KYQQospM8esqQoBLSKYBiwlgy30xw5V2Hlfb7YCshNJCCCGEEEJMIxIACiHEBCnFK6uJ4BEBjEy4FFVli/1Qk+Z/7XZc5C8C8iiJooUQQgghhJhWJAAUQogJEkyvLNBrYrxsY8V1AKXqSlSHQRWDvjS/s9txqr8bKIOSSlQhhBBCCCGmHQkAhRBiAjkYUBGWEwWMBC9i3Nli+KfJAwUuNYt5r9m9GP750geFEEIIIYSYhiQAFEKICWLLJ2HLEzYBsuqaGOf+FlT9+WiGWUmCvc3+fNMuAAoS/gkhhBBCCDGNSQAohBATKFjzz+dRagGNKxGMGAdecbqvJgNYvmYXMcc7iEdsC4o0yLRfIYQQQgghpjVXmkAIISaOQQOG+0wDaKn/E5XlAwqFSw6w/NW2c45dyL9tE6gcDtniWpRCCCGEEEKI6UwCQCGEmHAeq3QNljAKD4sjU4HFFvYohYPFoQB4PPn/7d17jGxbfhf279pV3ed1Z+bOnbkzd57MeDz2+BGPEQ62BwLE4CQKOFFQIoJEIiFFAsQfRCjhFWJFyoPHQBBEsTEEghJHMRKCPIyFSZBAYPsPEgUDxg88gPEw4xnPzJ07955Hd1XtlT/23lW7qqv79Hnd073r85HO6e6qXbuqV1Xv2utbv7VWfUd+9+pj+RvllX75mXup/WBgAABg+pz5AzxH62GXbcnfrm9Lskor/uMxX0urfp6/eU5S8iA/nxfym9t/OR9v/5X8jfKuJA/S5CRtv/4vAABwGFQAAjxnTZK2lPztvJhflc+kzY3MNAuPYJGSo7SZZZlkmZ/OS/mv2o/k+9v3JaVJssgsy6zS9MPOueru3r2bpvFcPU1t2+bOnTvX5vHev38/bdt6HTzl18Dt27dTig/aADg8AkCA52yWNm1K/lZ9W/6zUsUzXMqwqu8sbY7yIEnJz9a35T+vH80PtK8k5SgpJ2mySk1juO810rZtvvjFLyaJoOJp/b3Urub15s2bmc1m1+I18Oqrr2axWAgAn/Jr4MaNG5nPdYEAODze/QCed0cvSbLM32vfktq8kFkf2Oj2s88qJW2So6z6Of5K/np9T/7U6oP5q+U9SUpSFpnnXpYq/q6lUkrm87nw7ymrtV6bNi2lpGmazOdzAeBTfg0AwKESAAI8Z11As8ov5oX8g3on31S+mFVuZm6WNoZOax8Ul5TMsuiDv+P8xfrB/BerD+enyov9UN8HadKmpslS8He9n/NrFFYBAHD1CQABnndHP8lR2izKPD+Wt+ab8osahWS0qMdmfr9FPlvflj+fj+RPrT6Qz5e3JqVNcpp5WhV/AADAXgJAgCtglSRZ5q/Ul/Pbyj9V/XfA6npYeMksy8xymmSWf5C357vbD+d/a9+dlFtJOU3J/ZRUFX8AAMCFBIAAV0BXtbXID7cvJ81xHwEJdA5RSV1X+63yQv7H+oH8mfre/N325aSkn9/vjbT966aaLRIAAHgIASDAlVGTcpS/Xt+Rf618NqvczEwl4IFpk9T8k7w1f7x9f/5C+77cK7e7y8tJmj4YVu0HAAA8Cj0IgCti3oc7P1DfnWTZDwPlUHRR7yq/p35NPrL8tfme+pHcK8dpci/zPEj6ij+vCwAA4NH7mwBcCV0AtMxfy0tJbuUoq6zSZKZpDkLp1/n9U6v3J+Uk8yz6wE/oBwAAPBkVgABXxKqfB/Cz7dvyI/WlJKfmdzuY5777/yfztpzkVpJVv6IvAADAkxMAAlwhR6lJKfn++u5+FVhzAB6CNiXJMv9z++6kFNEfAADwVAkAAa6QbnGHk/y59j1JjvphoUzdPDXJLH8t70gSw74BAICnSgAIcIV09X5tFuVW/lJ9JclploYBT9oyJSWLfKa+kP+vfWuShVV+AQCAp0oPA+CKmfdVf3+yfqAfBqwKcPqW+Rt5e1JuZ56lgd+kFME/AABPs58JwJWy6ocB/532pXyqvJiPlNeyzI1+mCjTeyOuSZp8b31fklNxL0mStm2FgE9ZrTW11mvzWNu2Tds6IjztdgWAw+13AHC1OihJZmmzKjfzx+r7873ly8q1J/xcl6zyRr2VH2vfmZRFWs82SW7evCkAfNp/b7VemzYtpeTmzZuCYADgqREAAlxBXQh0mj/dfih/ovknuZnT1MzMBjgxq5TMc5rvy4eSMkuTpQpAUkrJO9/5Tg1x4K+Bl156SUMAAE+NMgOAK6irAlwm5TjfXT+U5CSt+G9yz3E33+M8f6G+J0m16jMAAPBMCAABrqjazwX4J1cfzFfqWzLLMish4GR0qzuf5O/Wl/IP23f08/95WwYAAJ4+PQ2AK6rtqwBPy+38vnx1klMH7Qk56r/+ifqBpHSrPZueHgAAeBb0JQGusG5F4Af53tUvyWfriylWiZ2ELug7zWv1Lflf2/cmOe2fawAAgKdPbwPgiptllZRZfms+1tcFVpVi115Jssh/mq9KynE33yMAAMAzIgAEuOKGKsAfXr0/f6l+IE0eWBDkGlumpOQ0X6xvy59dvd/cfwAAwDOnxwFwDZSUpJzmN62+ISe5k1mWqULAa2mWJFnmt+djSbnRP5cAAADPjgAQ4Bqo/YIgbbmVf7f9xiSLxKIR106bkpIH+bH67vyl1fuTPDD3HwAA8MzpdQBcE92w3/v5wfq+fLJ+TUruqwK8RpYpaXKa5DjftfqmpKzSWNIFAAB4E8w1AcD1UPuhwDWn+T2rr8+3z76cX1k+lza306gFvNLalMyzSrLKr2+/JV8sb03JfXP/ca67d++mFAH/dVNKSdu2KaXk9u3bj3/MaNucnJykbds0TZNaHeOv3Xt2rblz546GAODKEAACXKcORZImq7Rlll+3+pZ8av4jeV++kmVuZi4EvJLP1yol85wmWeW3tr80P1Tfl6je5AJt2+ZLX/qSAPAaP39HR0e5devWEz2Hr776apbLZZrGBwXX8vgvAATgihEAAly3zmWalCxyUo7zseW35x/P/k5eKXfT5qZKwKvS8eur/mZpM8/93M/t/HvtN+Wv1vcmOekrOWG/Ukpms5kA8Bo/f03TPNHzN+xjNpsJAK/re3VrigcArhZnFADXUO3nk3uj3Mw3r749X8qdNLmflWDpOT4nySKlX+hjlVkeJFnmz9WvyjvaX9OHfw+SVM8RD389GfIJAMBTJAAEuKbalDQ5yefK7Xxg+Svz/9Z3ZJa7KVkJAp+hOvq3Sln/K2lzlPtpci9JyV9sP5iPtr8q/9Hql+Z+nafkQWLYLwAA8BwYAgxwjbV9JeC9ciPf0n4if6j8TH5f87N99dlxlmnMDfgYllsBakmbZJaa2XrmvjZJm1n/NVkluZ3/u76S/72+O38+r+RevZNkmZT7/ZBf4R8AAPB8CAABrrmhErDNPL+/fkP+wvKV/JnZT+dXlV/oD/I3skqTRgT1UN1KyzXzLPpgb1zvV/p/sySznOQ4P15fyN/Mi/mh+s78ePuWfLnc7N9aT1JyPyU1rZkZAeDqnkf1K3dzwOd/tZpvlYMgAASYwslrmpS0qbmfny4v5le3n8i/WT6TP1D+aX5F+VxmKUmOsupr1pq+oo1xG5Y0WSZp8+P17Xk189zLPK9nls/lKJ+rN/LpHOcf53Z+JrfyxfZOFwaWfqL3skqyyFFOsuzjVpErAFxtgh8EwBwKASDARHRx3lAN2OSH6nvzQ+278yuaL+S/bv5JfnW+kFnursPAZJ5lSh9R1a1JYcvBtV03lDpJfkv78fwv7Qf7YK/p/83Xw37XQ37LIiWLlCRN2r5esMnC9LoAcG381E/9VNq2FQQeqLZtc3R0lI9+9KMag8kTAAJM7USmD6BmuZdVafIj9eX8muW7857mtfyW8rl8e17Nr8xX8nJ5I/Os1kNaO13gtUrJsv9pM/B1ehWDQ2haci+v1hfyHfVb8vfqO5Jy2tcEduHebLT9EPSNBwi3Qj8AuJZ+8id/MovFQgB4oJbLZW7fvi0A5CAIAAEmapWmD+8epC1NPlvv5JP1a7u4qp7ko83d/Lq8mk+U1/LLy2v5cB5klm4Y62w9WHiY867JKkeTmkdwldL/jg/yo/Vd+c7VL8u9cisl97dqIFdpsvJyAoBJunnzZo6OjgSAB2q5XObWrVsagoMgAASYsDoa3NtkmVlOs0yTtjT56fpifjov5b+vSWpJsshHmjfytbmbD5f7+XAe5P05yXtzkm8s9/L2vN7PIzi/9kHgMiXzLJMs84frx/L7V1+TlJJ5HmRp3j4AOBgPHjzIyclJZrOZxjhAy+VSI3AwBIAAB6JdLxbSDWttsky3fEWTtnRfP1Xfkk/lxaQOg3/bpC4zzyq/s/m5/LHmZzPPvSTHaTO/dstc1H6+v3nu5X69k3+7/rL8X/W9STlJk1WWhvICwEH5uq/7utRaLQRxqOfH/RyAcAgEgAAH5rx560ralL7+bYj/VklWpcky8/zJ+rX5nvYD+S/Lp/J7yz9Pk3tJbmSRWebXIAhcpOQoy5Sc5m/W9+Q3rj6eV8vtJPdTUszjBwAH6GMf+5hGAA6CABCAZGe48HjOu24ewWVqTrOo8/y++i/lD9UP5b+b/XT+g/LpHOU0yY2s0lzJocHdMiclR3mQ1Hl+Z/3mfE/9qqQs0+RB2pQJLm/CdacSBeDN07at4+6hnv/Wav5HDoYAEICLT4zWq+U2fZXgvbxWbuY/bL8lf7h8OH+0/OP8+vLZfpXg4yzS5OgKRGq1X8d3lmWS0/yd+p785vqN+XR9S5LTNFmp+kNnlGfy3LVt+8Qd0qexH57v64DrQwB0uLzXckgEgABcvlPaz6HX5DRtFvlH9cX8hvZb8w3NF/JHyqfy68sv5Cir51oRuHmMbWa5nwe5k9/WfmP+p/aDSUmaPEg15Jcr7saNGxrhuh4na818Pn+iOcVqrTk+Pk6t6pMBgKdDAAjAI2vXKws/SFua/ER9Kb+hfTkfb34xf6T8bP718pnMUpIcZZFZmtQ867X1ar+671HalJwkafI99aP5XfWjWdZb64U+BH9cdaWUvPzyyxrigM1ms7zzne/UEADAUyMABOCxDasKlz4I/PH6Uv6N9pfn65sv5bvLP8tvKp/NUb9q8Crz9SyDT7MqcDPUt+3m+cssf6W+P//x6qP55+XFJMvMci9tv/YxAMD6XMaUCwfPPIAcCgEgAE920jRaPGSoCPxH9aX8++3L+d3Nl/N783P5Hc2nc5S7SY6SzNJmllXy2JWBqz70O0pNyTKzLJIc5f+oH8h31w/nx+vLSVmtV/hdCf4AgD0EPwiAORQCQACemnFFYC1NPlPv5Hfl4/kDy6/Ov9N8Lv9J+fl8vHwlTe71Wx6nzSztVlVg3YrrNtOod6v1liSzdehXsszNfF/7ofw3+SX5TPvWpNQk99OkTZvGCr8AwLk+97nPPdGcnVzzc9e2zWw2y7ve9S6NweQJAAF4qrYrAk+TnOZuOcr31w/l+9sP523NG/nN5XP5DfnF/KvltdzO62nWA4PPDhDeVAiu1gN+T/NCfqi+kr9Y35UfyCtJvdFdX07WwZ/hvgDAw/zoj/5oTk9PVQIeoFJKlstl7ty5k+/6ru/SIEyeABCAZ2azWMgiySJtafJaPc6frl+dP10/nFt5kK9q7uZX5Cv5tnwlX1/u5oN5kLeXZY5S0yY5Scln6418KrfzY/Wt+T/zcn6mvZm75U4fFp6myb2kDx4FfwDAZd28eTPz+Tyz2UxjHKDlcpmbN29qCA6CABCAZ24cypW0meduFqXJ/RzlJ+o78hN5Z/5MktSS1GGg76q/RX9CXobBvDUpq5ScZJ42S6EfAPCYTk5OslgsMpvNUquJQw7JUAEo/OVQCAABeFPVJIv1XIGrNFn08/z18/WVbM3bV7JY37L0W7V9td9C8AcAPIH3vOc9WS6X5gA8QKWUrFYrFYAcDAEgAM9FXc/o1zx0u/40rd/eCToA8HR867d+q0YADoLSCQAAAACYMAEgAAAAAEyYABAAAAAAJkwACAAAAAATJgAEAAAAgAkTAAIAAADAhAkAAQCumFqrRrjGz93TeP6e1n7wNwwASTLXBAAAV0fbtvnsZz+rIa6pWmtms1leeeWVlFIe+zXw+c9/Psvl8rH3wfP3vve9TyMAcGUIAAEArpAh8GkaAzWuo7Zt0zTNEwV3w21LKV4H1/h1AABXiTMKAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAK6YUopGAADgqZlrAgCAq6VtWyHgNX7u2rZ9on3UWp/Kfni+rwMAuEoEgAAAV8yNGzfSNAZqXEdt22Y+n6fW+tghbiklN2/ezGq18jq4xq8DALhKBIAAAFdIKSUvv/yyhjjw18BLL72kIeBN8CRhPdPQtq0PWzgIAkAAAAAO0g//8A9nsVgIAQ9QKSWr1Sq3bt3Kd37nd2oQJk8ACAAAwEFaLBaG2x+w5XKZ5XKpITgIjnIAAAAcJJV/wKFQAQgAAMBBapomtdbUWjXGAb8G4BAIAAEAADhI3/Ed35GmaVQCHqi2bYW/HAwBIAAAAAfp5s2bGgE4CGpdAQAAAGDCBIAAAAAAMGECQAAAAACYMAEgAAAAAEyYABAAAAAAJkwACAAAAAATNtcEAAAAHKIf+ZEfyXK5TNOojTlEbdvmxo0b+bZv+zaNweQJAAEAADhIX/jCF7JYLASAB2q5XObOnTsagoMgAAQAAOAwO8TzeZqmEQAeoFJKlstl5nOxCAdyvNMEAAAAHKLlcqkC8MCf/9lspiE4CAJAAAAADtInPvGJpK8G4/C0bSv85WAIAAEAADhIL7/8skYADoIAEAAAgIPUtq3qvwNXa1UFyEEQAAIAAHCQBD8IgDmY450mAAAAAIDpEgACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmLC5JgAApq7WmiRZrVZpmmb9MwDAdTiPmc/nzl94IgJAAGDyXnzxxSTJjRs3NAYAcC298MILGoHHJgAEACbvB3/wB/P666/n9ddfz2w20yAAwLXRtm3e+ta35id+4ieS0cgGeBQCQABg8j75yU/mk5/8pIYAAK69tm01Ao9MAAgATF4pRSMAANee6j8elwAQAHCyDAAAE9ZoAgAAAACYLgEgAAAAAEyYABAAAAAAJkwACAAAAAATJgAEAAAAgAkTAAIAAADAhAkAAQAAAGDCBIAAAAAAMGECQAAAAACYMAEgAAAAAEyYABAAAAAAJkwACAAAAAATJgAEAAAAgAmbawIA4PopmgCulKoJHJc1AVxJTZKVZkAACABcR8IGAMdl4OFaTUBiCDAAcC2sFsnde/15rE+xAQAudw7Vnzc9eOAc6sCV/OUf81ENAHC13biR/PzPJQ/uJTduKjSBq6JdJm97KflbP5T8D39CexxUT7IktT8Y/8H/NvnoNySvfzlpDDKDq/W3muTe3eTd70lefCk5Oen+fjk4js4AwNV3epp86CPJ0fGmwwk8f8tl8q5Xkp/5h9rikH3oo8nXf3Pyxc8nc11MuHJKkgf3hX8HztEZALj6ak3e+ErSCv/gSnUol6su8Ln3hvY4ZG+8lnzpC8mrX0zmM1XacBU1TfePgyUABACuyYnrzOzFcBXNZjqVjs/d62A+S2a6mABX8lCtCQAAgMei0guvB4BrQQAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAAMCECQABAAAAYMIEgAAAAAAwYQJAAAAAAJgwASAAAAAATJgAEAAAAAAmTAAIAAAAABMmAAQAAACACRMAAgAAj6doArweAK6DuSYAAGCSVqukVu3wrJQky1XXzm2rPQ5Z278OlqvuZ392z05TkmamHYBHJgAEAGCCZ7lHyUsvJ8c3VCU9S6tV8tK7khffoS0O2TvenbzvQ8mdO8lMOPXs/t5qcno/ef0ryWqpPYBHOzXSBAAATEopSWmS3/Ebkwf3kzJL0hqf+CzUmty4mXz5i9riEJ/7wR/87cmNW8nitPv742k3dlf1d3qafPXXJb//jyavvZo0ZvQCLk8ACADAtDRNcjRPfvLvawt4M/zcp7TBm2XxILlxQzsAj0wACADA9DRNV5l2cpLM5kk1R90zVas2PvS/NxW2z9ZsnixOkre9Q5Ul8FgEgABwFZlQ/xHUna9v1nP0vDraV+l1+ibfX9n5etF2q2WyXHavi3ZpMRDwnjUN5v4DHpMAEIAn7/W3E/gdHtk5KUxTdtKJ9vxtz/sEv5SHz+tz7qf/5SHXb2927pVNebzf/zE3u+QD2799eYzbPc5tLv0cPGNXKch6Hm3wsN+/1mQ+76r/ZjOdZWBiVP8Bj0cACByWZ/IJdbv/oib9f+2jpSDrm1y21OVxO+Rl9KVc4rzynO2H+ywP+6V29n/mcV6w/+xrwubybVHOeUzlku20LwxbtUld7b/8TCDRJu1q85hLkrb2l533slpe/NwvFhc85rr/cWxdf8F913rx30qt+3/3vW30iGFVu3q0Sr46tPmj3s/y6RxP2vrm98Xa2lW2Pc8+YE0XsjXlzb/fpnn4/Taz5OimqiQAgJ4AEHiavdKtL89s/2d7ejvf7gZDfTi1VVU1qhoq+8KvZrSbZs9djcK5M7fvLxsHH8P3w/XnXVdr18EtQ54xXFe6juxwX7uhyr79Db3lOux8fPvRvsff79tH2/YVNMNt22S52r6P5c7wunYx2mXpbj8Ou1bt6OfShVHjQGl9H2Xz3G8FXiVZnW4HZDXJ8vRsaLa+70tuu69tl6dnX1PtvnCrduHMmdCiPXt5ecgwnrZeHGq1Dwn4lqvzA8SSi8PHSwWAlwjc6mp/e5yn5CHB5UWP5RFDw7p68iqK1er5VONdmVUf2+dT/Tsr/aq+FzzBpekmyV8u9h8zAQAOjAAQrm0HamfOq/a88Gunc31h1dN5VVw7Qwr3bVOGYY9l/WXUWx19KTv3VbY22a4UK9uBXNl5TEMFyGqZ1D7gatu+c98HGGXcSR+FGrXdVEC1ffVQrZtKonUA0ldKlZ1qpqFKqY4qreqqexzDvtvVZkL0ceBV+8BsMHRQm9Jdvlp1HfzVqvvdStMHYotNey37+y/pb9NuXg/ryq66mf8qGYV0TVf9NGw3PN6hfdePtey81va8/NZPR83ZF8fost3Od+mfr5Rz72LrwpKk3X19XuJxZXQ/5WHbnndh2X/xuX9qzZNtm5z/d3apv9ds/70+3s4vDplKuVylafMmnWaUxx02/GbfL09FzcWBb+2Pp0dH3etYFSAwJU3ppjdoZpv36vqQD+0uvK5ecEhtd/oZeYTJWM877Wk8h/CcCADhQk9S3dDu6XCX7Y51KUmZd++DQ4A2Dr3WHfDdAG1cvZazFWe1nv05dfskYO8245OE2oVZZVTps1U1VjcnHMvVJtSqfbVYGQVJQxg2/rm2m/CsrpJFv49hovblsrt8CLpWqz7cG4VrQ1XaOMAbKujSJqu6GWo5VPuMQ7jhd1i13e/ZZlOFNg4AhwqmoZppXKU0XvWwHZ1ErZbnByh7Lx9VKW5dtqe6cF/4UMoltyn7s6fxY5o3228PV+Vk7bHDlvJIFz+7wOKRrrgiIcsTbfAmPtb66L+cgrCJv30XwR8wMaNRIKenXRA4nPc1s8255HD+fuYD/NHXoR+wr08xVFGvtxndZnj/HPcXUkfdnqHv1I6urpvz9OyMcGjb7f1snzA/XjMJGeFcAkCu+hn8JQK4nWGnTXP5N4LdsCWjeddK2T8MdAjhxmHKeP/rAG9UsdaUzTxV7WozpHIImVZtUpf9fFLLTTDWtpuqsvX2q03oNFSk1VUXdrV9ALdc9nNELfp9Dfse3e9yufPzYjMn2Wq0n+G6mk1F21aF2G6VVt0JoUYd7VL2dLzrTnXXOGwsm+uzW/UxCi+35oJq9pwzlJ355bJdAVT2fJK5fk5HlYZ7q5nKQ/Z52dyknm3Ph97uEROMesn7ubKB1OM+5nrl8ysAgItPZ/oPqX/hM8mf/ePJvdeT2VE/xc2smx6hmXc/z+f9fKnzvlKwJPOjzXWl2WwzfD8rSdNvM+tvN++vb+bJrNncVyn9z7PNFDvDvoZ5WkvTVSoOl836isV5X704G7Yfvs/oQ/e+c7cVNo77haPpP9rdPuFO4LjOKttLnFee0/ds2/2jlfYRQHKFCQB5xi5bQdduBzVDiNf0b0brEKbZHFSbnYq50owO5nUTdtQ9FXB15/vx3GupyWKZrBZd8LXsq82Wyy5gW676r32AtlyMvh8uH20/DOMcKt3ancq3tg/zhjnChsuHcHBd1dZftqqb6rnx11W7HX5uvfk022X7Zc/8d+PLtqrRdkLO0uypEHsKb3IPW9E0j1ixVZ9C0lMvc/u6vd21CJUkXwAA17MHP+/CvCH8Sz/38jJJTkejYrKprKs5OzKojkKxuhuQ7fTh2nY7ABz6a8OUIEMoWPrgr+nnap2Nfx5CwnGAWDaXzZrud5o1XQA5m3dh4exoM+R5Put+92YIMIdt5n2IOetCzPk8OZpvHldGj7HuTGWyNa3Jbj9pdFkdB5KjSsf113p2yqFhNNH63PsSfSYBIs/y8KEJOOu80K6/fCtkavaHOKX/FKccbYK7IVAqzfaBbXhDGA66Q2i2DtIWXbi1WozCt9WmWm7Vf132wzgXi+7famc/i5P+8j6QG/a13r6/n+Vie5jo7qILW4s71M2byHixhVJ3hsvuzHM3tM/WvHY7QebQvus31vmett59g3qEAOwy4dbD5ll6qplUfXhYJbcCAODQNc3O4na7H/ZfdOPz5vQuF/64Ka6oZ8/dt6ZcWPVh5DJZ7ISR41ByPFR4XQtSt3+HOh4NVDaFGyln+2Xr/tjOiJx1hWMfJh71/8bh4WzWX36jDxqH7efd5cfHfTVlswkj14Flv30z23x/dLzZx/xoM0f5EBRuTTE0XDaqYKzjdmkf0t5ttjua49dJhIlsEQAeor1z4vQzuzajEu9mVM6d0Sc3w6ca40+K2tGnR8NCDItFcnLSrZ65WCTLk66y7uSkC9kWw+Wn3TwWy0X3tV1t/q1Wm7nfhiCvHc8nN/q36ueKa3YDxmFobrMJ2Maf5gwB5bDt8GnSpQ+WZf8b6EVvvJeqLHvIdg+dtBcAADgsD+sf1Et1Qx5JMyrmKLNk9jj9p30Psz78svM7vdtFLUPRx4Odar2hT7uvb7veR90surIe0txsh4KzPd83s64ScXbcB4zHm6DxaN5/PepCwhvH3XbHR8mNW5t++NAvHYpp1qPgRtNPrQPFcf+4DxKHn1P3D6dqBISHRAA4KTvzIoyH0pay/UlRM5qPoZlt5lw47QO7k/tdQLdcdT8vl/3Pyz6oO0ke3O/CvNP7fYC36rY5ud/ffjGabqxuTztWs6mSK+NPlMpoiG//texcVpr+gJntoarjIaTnVb+d96ZRn+QdsO6fm00wBwAAHIzL9oHejL5Ss3/Bu73TDu0GknumIKp1eyqpIXSrq+R0ldQH5w8RLvXsIn3jkWbZt5BfuoDw+EYXCB4fJce3kps3kuObyY3+uuNbfd/4uAsS5/0Q6OPj7rKbt5L5cT8nfR8Mtu1mbvdxWFjHi7PUswtYcu0JAK+7cTXfvC9RHiZ0rf0qqHW0cMODe8ndu92ksffeSO6+0YV19+8lpw82Id7pMCR2GIK72gzB3ZrkdTyvQxlVETbdwem8KrozK53uf184f+jqzhDcKm0DAADgAnv7jY8TSI7nL9xz3cP6txdWM44Kex7cT+7d3QR0bXu2sq8ZhiMfbYY1D7nA0XEXAB7fSG7fSW7e7L7eeWv39fYLXZjYHI1GAPbzJQ7TcS37KbQGAsFrSwB4nTWzLv0/PuoOMne/krz2pe7rG693/15/LXnjte77u29s5rYbH1TKaK67cVnzcFC7cXPzh76eD++Cg9dlwritbVTLAQAAcJ2c14990v7tqHpx3AdPdhYsGfWtxwuPLBfd/Pd1NAz4TMVhs+nLHx8lt9+SvOWtyQvDv7cld+4kd4bL376Z5msoFOLaEQBeV8NcfJ//dPLPfy75/L/oKvru3Uvu3+3+4JvxhKT9ykjN8XgnZxeQOBPmjX5uW+0OAAAAb5b6CMU2w2IkZ/r8exZyGS5sa1c09JUvdQtrDoty3uirBW/dTt7yYvLu9yevvLcLBI9vyAeuIQHgddO2XTnv/XvJD/5A8vP/rBvX3zRd6e7sKLnzQjJ/sd8+m8lAc17lnQo8AAAAuN7q2fnwL+zrN90Q5vlRHxT21Ydtukq/xSK5+/nk0/8s+fv/Txf+ffO3Jr/sE5r6GhIAXjdN06XxN28mv/bfSn7h08ln/0XyhV/o5gdoV8m9fiXdmn558tlmhaJhTP/WweASi2QAAAAA18R4LsI9C53UtssW1isH93P91dGqx03TZQ9f9TXJV31t8vJ7kre8rVsclGtHAHhdtTV51yvJK+9PPt7/wX7ly8nrX06+8nry+pe6ef8e3Evu3U8W95OT036xj5PRasD9xKXD6rrNsJBHGa36k+3FPNYr7o6PIMJEAAAAeHz7FhDZmbprPJT3vCm72rYfCVjPrvY7zAk4m3ejCG/c7Fcbvtn9u3kneetbunkAb78leec7u6+lJItlN90Y15IA8Dq7f7/7OgR5L7wlefHtXbXfbN4dHE4edP8Wp8npaTch6OlJN1fgvTe6+QLvv5Hcf9Bdf3IvOTnpxv3X0UEi2UwimmwuL2X7MWyFiOPFRPYEhme+FR4CAABwnZ0T4iVnq/CGzu+wou96dd+hL163F/MY+uWlXxxkKNYpfUFP6Yfwzo66yr2bt5Kbt/t//c83bnY/37jdLQByfNytFnx83C0yOp/3lYHLbk7A5bIrJBru3yrA15YA8Dob/+G1/WSdp6fDBX3wNuv+3b7TBYTDQaHszAUwzA1Ya3fwWC67IPD0JDm9nzw43Xx/etpd9+B+sjzttl2edvMDLJb9XAGnXfnwgwf9gWPZPcYhCBwe+zowzKbysJRNiJg9FYj7fv/1MXR3VaTRgfZxgkZBJAAAwDWzL4Q777o9hSrjfuCZhTfafq798y4bhXepm3n5a7ogb3fV3qbpArjZrJuL76j/N5t3Pw+Xz466r/N5X7V33Ffw7YZ7t7rr6s7vWoYKwnHwOBoCvFwmJ6/1mUAd9ceb8/vfXCsCwEkZLRc+DsyWi2R5/rFvvQR4xgeG0pcB30qaF/s/9qGqr/TBXdMdVJbLzVLgy0UX/A3fL4dPDlZJ208iejqEhSf919PkdNGHiafdzyf99e2q26au+tCybD4FGULM4cA6HJfXPw//+vSvKTsHruGTktGQ5nF7NOM2bbbb5rxqxr0XnRMgChcBAICD85Bwbs+3F17f1u3VbddDYdvRopjZ6SMm68RuXVnX/zf0+WrZhGZbl422KaW7XUkyP+7CufnxJsSbz5OjG8nRvA/1biQ35l2l3dFx1+cegr3ZbDN//1FfyDOf9yv7HiXz0Qq/7ShIXFcMjv6dnJwNLh+2au96BB9TJQA8BI/6RzwcFMerB5/RH0TXIdlQgjzrPqloytnhv0N13jioGyrzyrgKcfRz6ScnXSy6oHGx6ILE02UfMPah4mq52WbVX77+ftUFiKvx5Kb9QXLZD3VetaNPP5b9z/0nIqf95bWvYhz2Mx7WXIYwsW+PMn5TG0LE3aAxe+ZWPPed7nLPZSkXvLGWnUrI9RO+59tHCCYFmQBcVbXVBsBEjmdDuFU387mX8pAbXTTs9BFHSdWLzvf3VcSNdlZ3V6LdCeTWc9glWzvaDfbq+Lhe++v7tpgPc9v3i14OQdp8WBRzCNP6EXHzeb8Cbr/9vEma4y6oa3Yr8Y66y+dHfZg3S2Z92Hd03PV/j46T2RAGlu0GLmXTt03ZFLYMl4+DybZuKgiHyxd9kUzb9hWE2V/4kwtGz0UFHwJAHubcg0Rz9rhSaxe4PfJ9lPNDr+HgWUp38D0+3q7K2/26nvtg9InMuBPQtsmqdoHg8OnIePWjYd7Dtt2EhnUUHNb+9kOwuK5+HKoel5vwsR1dP/w8DJEeSqzbdjO/wqrdXkxlqFxc/5y+FLtsD9nevU2p2xPDrrfrw9Txc7r1/DZny8NTtk8cxqHl7pvL1r52JqodPhnbd0Lw0JOdx7jNmU0FlAAHZ7Uada4AJqL21V2LRd/XGYaW5uyQ1fXX3cUhhu/rJpDaHSo6Pp8fCjfGRR1lHG7tnPtnFG41fT9gXck2VLM1feHIvKtsW18370K5+XBdH9oN2w1DYIfhsUMYOhvmv5ttRquVWTLLZlqspg8DS7O9IOasbILBjApidofqZrS4xvD9EEouT7tRd/Wi0V/to3dLxn2uYVQePKaSv/xjzoq4Ri76dGm0TXJ2u33B1+7w34zKvHeHVJedQHII5IY5FMqe0G4dfO1UN2YUypXhk7x+uPY6FFz11Yt9OLhabqoPl8tRVWM/OeuwcMs6WBzNvVjbTSC5rnism3kfhje32v+c2u2z1O6xjN8E1z/329dVH6q2m0+t6mqzUMx6n2U03+P4k8dmO//dGmJd9rzHNZtdbC1rv7PvZs9t1ldfotryvDfeRzrClnMPvfu+Pffyc4/S9YIfn9GhXcUncF3MZt3BdFgYDa5/z+1xr3yM7S5zbnSZfdUzp+gP375sNq6X2X3df3+XPX9az9E2eqC7uxxXqW39TnV/5VwdB0F1O7w7c/ejfYwLGZpZd928r9mZz7vz9dKMpkQqfaBVRlWBO9M2ldKHXMPc67PNvOuzoWpuCMHm3eXzPoQr/ciu2XwTls37qrpmvhmyOq62K0Ogd9Ttcwj8xn2lrZFhQwA5OkevO4FkLUkz7neNK+rG7bo77162hwFvrZ573sq5j3D+/7h9BHhOVAByzTSXOMY2ewKg897/h7Dq6T/My78Z7E7M2lc7luM9v/N51XejNGz9/c58FcMNax/grUYBX8omCKyjYLAtm+2HN9ZhharVqj8/G69QtdouX687IeCqr6JctpuwstZuWHet3c/J5vJau1A06QLOtn+DXyxHl9VR9Wnpth9+h/X+FpvHvlpm9DHp6MSxnD/Z7/D9+lPQdvukd9+2W9+XpLRn7+PC25bRYj7lEq/xfUHqeT+Xvbs79w/nouECe6/bDTLHVaTl4nPwy3QgnlR9k+/vUh0W4KkbKgDblba46glWeYr7umjz+vRzsKd34rjnZPSi+brq7nlMPX/URLtnAYO9odxOYFUveM8aV5TVc8Ks8flHzaYSbD1+sW6fF5Q95w/r29XtJ/Lc240+jG/GI1DK+fdR+uCrqV1oVUaBVqldlVmyqTYbwrSS7pw96W+XTVA2hGbD7YYAr4yv64O2Ouy7DwBnw3l8P9fbrJ9WadYHguOvpZwd/VT6y2dlNE1RswkLa9k8/vFc8Fvh2Ph1NX5t7Sx0kZ1hu1vf97dftcnqwSOe/5z32n0G/bRzR0eBABAONFR8xklA3RMAnXmTfRTnndg9bOjuuOS/r74r83PaajeEKmeH/mY8bLiOKiDHgVv/c0kXOJbs+SS1D8PWJ5H17Fl9bXfCpbp90jeutMwoFByGgK8vGwLMnfBwHVCW7dsM3w/7H8LJMgogx+FmRqFmzeb7ZDOEfuvy0oWWw/6HeUFSNqFo6cPQ4VPRoYpz3IFo2007DFWc+zoYQ5XnuB3Pu76MA9fRvupq83pYrbY7EuthE6MT8fKIHbp9ixdl57m+aH9NudxxYffm5ZzO54Xz75SHnHy259/33t1d1Hkuz/Uw+pQ2mm7ooQLtMd4e69XYx5umffofSD6t85QLr6qX3vzcX7B9xPveDZkedpOah5ef1X2vl/rwwKLNJUvb9p3vjYOy2c77Ux/WjN9Dh0qwtN35WRl9wNbMtkeqjN8nb8y330fG2w1hUvbso5ltQqDx/TXpwrGMbjf+sG9YpKCMwq9aNo9/1leP1VW3WELpzxPXAduw/4zmasumQq22m+qz1M384+thpmX7/sqsq0gbvp/1j3UI63bPT/eFl7uh43ibWs853g9DYMfns9mcJ2a8Gms2c5+fCXl3Xi8XBXDjv4/VcE5Wzw+V94XMD/tg+FEDtsd937z0ORsgAISDDiebx3/frjXJKrkKBRNlz4nUU+t8j1Z9Puo/WT0+vqD9msudGK2HxOx8wrw7RHlcoXneSePwtR0t471qN1etlqNtVjuflo5O9lar0ZCJbIdyOScAXD+udjs4HYZ7j7evq80vX1fbFZLD3Jbr+9657fJ0u63Wi/Tk/MtK6cLS3cn9h8rRh162E1rWcVsOl40qTccbLpfbz2tbz952/Fy1iz2v27KZC3Tfa3p5zlyqbbupdN33d1seMQCp5eGfwl/2k/rLbneZoPKygc2jBjvPMwgyB92TvQc88X6eUgfySffz0NvXS/wZXTJMvtRj3fOh277ziXJR0HZOIDKEPmd6HudcfjTfv6+hYmrXbLYJbrYebj//15ntm+4+xm8bs1GF1XmPu5lvwqLdQGu9TT98so7OA2ZHZ+9/63GVTeXYOtCbnT0nmO1UZw3BWur29sP5TDOqmEvfPk3dhGm7AeAQgq0f+850Lc3ulC1l+zkYV+hthY+zc6oJ68XB1pmFIur+4bnDB2i7QdiZH/eFaaP72/d++8TH6ysS5J+7QMRDfgYmTQAI7D9rmNT5wDnDYuo5J27JMz55u6BCc18HuDziSl67YeNuld3RfP9Z4p4pGs+cPDZ7e+g7q8yVc/az53fZPdHe1xmt5bye5/bt2n2d3vrQm24FlONvLtUJ2DMk/DKZxYX7rvuf0/aC/a6WjxEoXmLhpuXy/M7/1n6WD//Fh7lNH9aeq+Xl2n65eLRs6HT5/IK4dmHE92O9FTXbAc3jdoKH6qAneAs5N1i6rGGerYteCE2zGTJ40bHhMr9PM394CDisinnhn2Q5/zE151y3Ht1Zz3lTu+h96zKHxbqpGLu0nbCznvN4Si5+zEOgVh9yozPvZXsWQzvzO7cXVC/unrfsm4vunPOMbM98cva68QdItfugbx2QXVT52J5/jtNe4u/y4gue2unsM93/QZxDA1MiAATYd7bYvAn38azup9adT71zRYeVPa+nuFy+U/o8PFaV0QWVPEdHF9zmxqP20h5rk0fa8Gnv78zmz+O5NoTpCQ5oT377p3b8a9+Em7dPaV/tJZv2CaqAV+3OlBGXfX+6wq+t9sCT+kcZ+nlu5ZgADOCqEgACTP8s3sn4dfJYHeT6mDlJ+8QZAnD13wKeYKMJe9w5YgHgehIAAsDBap7oagAAYBJn/gAAAADAdSYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJEwACAAAAwIQJAAEAAABgwgSAAAAAADBhAkAAAAAAmDABIAAAAABMmAAQAAAAACZMAAgAAAAAEyYABAAAAIAJa1KSlKIlAAAAAGBK+syvSc33pdbaX1y1DAAAAABca13G12V+3/f/A/dOA3WUs54KAAAAAElFTkSuQmCC)"
      ],
      "metadata": {
        "id": "37zvNvxD65G5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### **Objective**"
      ],
      "metadata": {
        "id": "Z3rKXgMZdSbq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Develop a Generative AI application using a Large Language Model to **automate the classification and processing of support tickets.** The application will aim to predict ticket categories, assign priority, suggest estimated resolution times, generate responses based on sentiment analysis, and store the results in a structured DataFrame.\n"
      ],
      "metadata": {
        "id": "trDwCVXRy-rn"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Sample of Expected Output**"
      ],
      "metadata": {
        "id": "DXtapml31CHh"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![image.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "cGHvpK77dCRp"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Please read the instructions carefully before starting the project.**\n",
        "\n",
        "This is a commented Python Notebook file in which all the instructions and tasks to be performed are mentioned.\n",
        "* Blanks '_______' are provided in the notebook that\n",
        "needs to be filled with an appropriate code to get the correct result. With every '_______' blank, there is a comment that briefly describes what needs to be filled in the blank space.\n",
        "* Identify the task to be performed correctly, and only then proceed to write the required code.\n",
        "* Please run the codes in a sequential manner from the beginning to avoid any unnecessary errors.\n",
        "* Add the results/observations (wherever mentioned) derived from the analysis in the presentation and submit the same. Any mathematical or computational details which are a graded part of the project can be included in the Appendix section of the presentation.\n"
      ],
      "metadata": {
        "id": "ANlIjNQ-d-QG"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Installing and Importing Necessary Libraries and Dependencies**"
      ],
      "metadata": {
        "id": "8vGSy1Cl1Xj9"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "m0rHx0ZC5Jbv",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0673f977-ddc9-4882-c59f-fb0430bbb713"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m36.7/36.7 MB\u001b[0m \u001b[31m108.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Installing build dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Getting requirements to build wheel ... \u001b[?25l\u001b[?25hdone\n",
            "  Installing backend dependencies ... \u001b[?25l\u001b[?25hdone\n",
            "  Preparing metadata (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m18.2/18.2 MB\u001b[0m \u001b[31m289.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m45.5/45.5 kB\u001b[0m \u001b[31m219.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m133.2/133.2 kB\u001b[0m \u001b[31m231.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Building wheel for llama-cpp-python (pyproject.toml) ... \u001b[?25l\u001b[?25hdone\n"
          ]
        }
      ],
      "source": [
        "# Installation for GPU llama-cpp-python\n",
        "!CMAKE_ARGS=\"-DLLAMA_CUBLAS=on\" FORCE_CMAKE=1 pip install llama-cpp-python==0.2.45 --force-reinstall --no-cache-dir -q"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# For downloading the models from HF Hub\n",
        "!pip install huggingface_hub==0.20.3 -q"
      ],
      "metadata": {
        "id": "j00mbD8t_CtZ"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Importing library for data manipulation\n",
        "import pandas as pd\n",
        "\n",
        "# Function to download the model from the Hugging Face model hub\n",
        "from huggingface_hub import hf_hub_download\n",
        "\n",
        "# Importing the Llama class from the llama_cpp module\n",
        "from llama_cpp import Llama\n",
        "\n",
        "# Importing the json module\n",
        "import json"
      ],
      "metadata": {
        "id": "ksv9hSCR4BM_"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Loading the Dataset**"
      ],
      "metadata": {
        "id": "oKTiay5Tv_SQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Complete the code to read the CSV file\n",
        "data = pd.read_csv(\"Support_ticket_text_data_mid_term.csv\")"
      ],
      "metadata": {
        "id": "QJmvO05qXOIq"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Data Overview**"
      ],
      "metadata": {
        "id": "Q4vPV6u7YWVd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# returns the first 5 rows\n",
        "data.head()\n"
      ],
      "metadata": {
        "id": "_uZIThVoYZ-e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "0341dd30-8ec9-4572-b496-49ebf47f5466"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text\n",
              "0      ST2023-006  My internet connection has significantly slowe...\n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...\n",
              "2      ST2023-008  I've accidentally deleted essential work docum...\n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...\n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve..."
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-facbb195-e091-45ba-b4af-095d5cab710f\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-facbb195-e091-45ba-b4af-095d5cab710f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-facbb195-e091-45ba-b4af-095d5cab710f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-facbb195-e091-45ba-b4af-095d5cab710f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-bbbb17bb-bd1b-4c07-94d9-80395311e86a\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-bbbb17bb-bd1b-4c07-94d9-80395311e86a')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-bbbb17bb-bd1b-4c07-94d9-80395311e86a button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data",
              "summary": "{\n  \"name\": \"data\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# get number of rows and cols of the dataset\n",
        "dataframe_shape = data.shape\n",
        "# get number of rows of the dataset\n",
        "num_rows = dataframe_shape[0]\n",
        "# get number of cols of the dataset\n",
        "num_columns = dataframe_shape[1]\n",
        "\n",
        "print(f\"Number of rows: {num_rows}\")\n",
        "print(f\"Number of columns: {num_columns}\")\n"
      ],
      "metadata": {
        "id": "UzSvf8sHYmNs",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "dc7cb89b-bd71-4a75-e46c-8c9afc75ba2d"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of rows: 21\n",
            "Number of columns: 2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# check missing data from the dataset by col wise\n",
        "data.isna().sum()\n"
      ],
      "metadata": {
        "id": "kRlXT2MnYg78",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "43cf9138-2e03-4bbc-9a3a-df6f25049093"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "support_tick_id        0\n",
              "support_ticket_text    0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Model Building**"
      ],
      "metadata": {
        "id": "TdDKo11ZZNOJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Loading the model"
      ],
      "metadata": {
        "id": "jbaduRVymY3v"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_name_or_path = \"TheBloke/Llama-2-13B-chat-GGUF\"\n",
        "model_basename = \"llama-2-13b-chat.Q5_K_M.gguf\" # the model is in gguf format"
      ],
      "metadata": {
        "id": "61N9i3i8K5_9"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Using hf_hub_download to download a model from the Hugging Face model hub\n",
        "# The repo_id parameter specifies the model name or path in the Hugging Face repository\n",
        "# The filename parameter specifies the name of the file to download\n",
        "model_path = hf_hub_download(\n",
        "    repo_id = model_name_or_path,\n",
        "    filename = model_basename\n",
        ")"
      ],
      "metadata": {
        "id": "0ZpjCaMRVcM-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173,
          "referenced_widgets": [
            "eb8e213c8d3d46759dfd29c498bd4ddc",
            "9fdf8f36d0d3434091558433f22adf1a",
            "5b8ca6845c5e4050aec4a1a2db4b5ca9",
            "47cd315e497242e6bd5f7126a5dea71f",
            "66045d939dac4e6d97aa0ee6dfc5a7ec",
            "be53f1f4896e4cc0a1f3c0df96e87c3e",
            "1ea44c1d53a24a2bbda3b2266ef05209",
            "484823f23fe4464d9aa51bce9e3fe17b",
            "ab7eda0f0b2e4dbb9d65ba9d6173d0d9",
            "7116a457d0e44940a601a5ce605291e0",
            "a1cf5d7e4b2e4ef9bc920de85d3e0556"
          ]
        },
        "outputId": "ad299377-68f5-451a-963a-91a4affb752e"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/huggingface_hub/utils/_token.py:88: UserWarning: \n",
            "The secret `HF_TOKEN` does not exist in your Colab secrets.\n",
            "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n",
            "You will be able to reuse this secret in all of your notebooks.\n",
            "Please note that authentication is recommended but still optional to access public models or datasets.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "llama-2-13b-chat.Q5_K_M.gguf:   0%|          | 0.00/9.23G [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "eb8e213c8d3d46759dfd29c498bd4ddc"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# complete the code to create an instance of the 'Llama' class with specified parameters\n",
        "\n",
        "lcpp_llm = Llama(\n",
        "    model_path=model_path,\n",
        "    n_threads=2,  # CPU cores\n",
        "    n_batch=512,  # Should be between 1 and n_ctx, consider the amount of VRAM in your GPU.\n",
        "    n_gpu_layers=43,  # Change this value based on your model and your GPU VRAM pool.\n",
        "    n_ctx=4096,  # Context window\n",
        ")"
      ],
      "metadata": {
        "id": "1SolgBhkVdgc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "dd34d29d-ed0c-4f7f-96b0-c2b63f760277"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "llama_model_loader: loaded meta data with 19 key-value pairs and 363 tensors from /root/.cache/huggingface/hub/models--TheBloke--Llama-2-13B-chat-GGUF/snapshots/4458acc949de0a9914c3eab623904d4fe999050a/llama-2-13b-chat.Q5_K_M.gguf (version GGUF V2)\n",
            "llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.\n",
            "llama_model_loader: - kv   0:                       general.architecture str              = llama\n",
            "llama_model_loader: - kv   1:                               general.name str              = LLaMA v2\n",
            "llama_model_loader: - kv   2:                       llama.context_length u32              = 4096\n",
            "llama_model_loader: - kv   3:                     llama.embedding_length u32              = 5120\n",
            "llama_model_loader: - kv   4:                          llama.block_count u32              = 40\n",
            "llama_model_loader: - kv   5:                  llama.feed_forward_length u32              = 13824\n",
            "llama_model_loader: - kv   6:                 llama.rope.dimension_count u32              = 128\n",
            "llama_model_loader: - kv   7:                 llama.attention.head_count u32              = 40\n",
            "llama_model_loader: - kv   8:              llama.attention.head_count_kv u32              = 40\n",
            "llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010\n",
            "llama_model_loader: - kv  10:                          general.file_type u32              = 17\n",
            "llama_model_loader: - kv  11:                       tokenizer.ggml.model str              = llama\n",
            "llama_model_loader: - kv  12:                      tokenizer.ggml.tokens arr[str,32000]   = [\"<unk>\", \"<s>\", \"</s>\", \"<0x00>\", \"<...\n",
            "llama_model_loader: - kv  13:                      tokenizer.ggml.scores arr[f32,32000]   = [0.000000, 0.000000, 0.000000, 0.0000...\n",
            "llama_model_loader: - kv  14:                  tokenizer.ggml.token_type arr[i32,32000]   = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ...\n",
            "llama_model_loader: - kv  15:                tokenizer.ggml.bos_token_id u32              = 1\n",
            "llama_model_loader: - kv  16:                tokenizer.ggml.eos_token_id u32              = 2\n",
            "llama_model_loader: - kv  17:            tokenizer.ggml.unknown_token_id u32              = 0\n",
            "llama_model_loader: - kv  18:               general.quantization_version u32              = 2\n",
            "llama_model_loader: - type  f32:   81 tensors\n",
            "llama_model_loader: - type q5_K:  241 tensors\n",
            "llama_model_loader: - type q6_K:   41 tensors\n",
            "llm_load_vocab: special tokens definition check successful ( 259/32000 ).\n",
            "llm_load_print_meta: format           = GGUF V2\n",
            "llm_load_print_meta: arch             = llama\n",
            "llm_load_print_meta: vocab type       = SPM\n",
            "llm_load_print_meta: n_vocab          = 32000\n",
            "llm_load_print_meta: n_merges         = 0\n",
            "llm_load_print_meta: n_ctx_train      = 4096\n",
            "llm_load_print_meta: n_embd           = 5120\n",
            "llm_load_print_meta: n_head           = 40\n",
            "llm_load_print_meta: n_head_kv        = 40\n",
            "llm_load_print_meta: n_layer          = 40\n",
            "llm_load_print_meta: n_rot            = 128\n",
            "llm_load_print_meta: n_embd_head_k    = 128\n",
            "llm_load_print_meta: n_embd_head_v    = 128\n",
            "llm_load_print_meta: n_gqa            = 1\n",
            "llm_load_print_meta: n_embd_k_gqa     = 5120\n",
            "llm_load_print_meta: n_embd_v_gqa     = 5120\n",
            "llm_load_print_meta: f_norm_eps       = 0.0e+00\n",
            "llm_load_print_meta: f_norm_rms_eps   = 1.0e-05\n",
            "llm_load_print_meta: f_clamp_kqv      = 0.0e+00\n",
            "llm_load_print_meta: f_max_alibi_bias = 0.0e+00\n",
            "llm_load_print_meta: n_ff             = 13824\n",
            "llm_load_print_meta: n_expert         = 0\n",
            "llm_load_print_meta: n_expert_used    = 0\n",
            "llm_load_print_meta: rope scaling     = linear\n",
            "llm_load_print_meta: freq_base_train  = 10000.0\n",
            "llm_load_print_meta: freq_scale_train = 1\n",
            "llm_load_print_meta: n_yarn_orig_ctx  = 4096\n",
            "llm_load_print_meta: rope_finetuned   = unknown\n",
            "llm_load_print_meta: model type       = 13B\n",
            "llm_load_print_meta: model ftype      = Q5_K - Medium\n",
            "llm_load_print_meta: model params     = 13.02 B\n",
            "llm_load_print_meta: model size       = 8.60 GiB (5.67 BPW) \n",
            "llm_load_print_meta: general.name     = LLaMA v2\n",
            "llm_load_print_meta: BOS token        = 1 '<s>'\n",
            "llm_load_print_meta: EOS token        = 2 '</s>'\n",
            "llm_load_print_meta: UNK token        = 0 '<unk>'\n",
            "llm_load_print_meta: LF token         = 13 '<0x0A>'\n",
            "llm_load_tensors: ggml ctx size =    0.28 MiB\n",
            "llm_load_tensors: offloading 40 repeating layers to GPU\n",
            "llm_load_tensors: offloading non-repeating layers to GPU\n",
            "llm_load_tensors: offloaded 41/41 layers to GPU\n",
            "llm_load_tensors:        CPU buffer size =   107.42 MiB\n",
            "llm_load_tensors:      CUDA0 buffer size =  8694.21 MiB\n",
            "...................................................................................................\n",
            "llama_new_context_with_model: n_ctx      = 4096\n",
            "llama_new_context_with_model: freq_base  = 10000.0\n",
            "llama_new_context_with_model: freq_scale = 1\n",
            "llama_kv_cache_init:      CUDA0 KV buffer size =  3200.00 MiB\n",
            "llama_new_context_with_model: KV self size  = 3200.00 MiB, K (f16): 1600.00 MiB, V (f16): 1600.00 MiB\n",
            "llama_new_context_with_model:  CUDA_Host input buffer size   =    19.04 MiB\n",
            "llama_new_context_with_model:      CUDA0 compute buffer size =   368.02 MiB\n",
            "llama_new_context_with_model:  CUDA_Host compute buffer size =    10.00 MiB\n",
            "llama_new_context_with_model: graph splits (measure): 3\n",
            "AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | \n",
            "Model metadata: {'tokenizer.ggml.unknown_token_id': '0', 'tokenizer.ggml.eos_token_id': '2', 'general.architecture': 'llama', 'llama.context_length': '4096', 'general.name': 'LLaMA v2', 'llama.embedding_length': '5120', 'llama.feed_forward_length': '13824', 'llama.attention.layer_norm_rms_epsilon': '0.000010', 'llama.rope.dimension_count': '128', 'llama.attention.head_count': '40', 'tokenizer.ggml.bos_token_id': '1', 'llama.block_count': '40', 'llama.attention.head_count_kv': '40', 'general.quantization_version': '2', 'tokenizer.ggml.model': 'llama', 'general.file_type': '17'}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Defining Model Response Parameters"
      ],
      "metadata": {
        "id": "VVXK7vYfmdkL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def generate_llama_response(instruction, review):\n",
        "\n",
        "    # System message explicitly instructing not to include the review text\n",
        "    system_message = \"\"\"\n",
        "        [INST]<<SYS>>\n",
        "        {}\n",
        "        <</SYS>>[/INST]\n",
        "    \"\"\".format(instruction)\n",
        "\n",
        "    # Combine user_prompt and system_message to create the prompt\n",
        "    prompt = f\"{review}\\n{system_message}\"\n",
        "\n",
        "    # Generate a response from the LLaMA model\n",
        "    response = lcpp_llm(\n",
        "        prompt=prompt,\n",
        "        max_tokens=1024,\n",
        "        temperature=0.01,\n",
        "        top_p=0.95,\n",
        "        repeat_penalty=1.2,\n",
        "        top_k=50,\n",
        "        stop=['INST'],\n",
        "        echo=False,\n",
        "        seed=42,\n",
        "    )\n",
        "\n",
        "    # Extract the sentiment from the response\n",
        "    response_text = response[\"choices\"][0][\"text\"]\n",
        "    return response_text"
      ],
      "metadata": {
        "id": "5AfP0mcXVfXV"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Task 1: Ticket Categorization**"
      ],
      "metadata": {
        "id": "ZvdWcnbphzuZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# create a copy of the data\n",
        "data_1 = data.copy()"
      ],
      "metadata": {
        "id": "t-nX7z_ih46S"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "instruction_1 = \"\"\"\n",
        "You are an AI analyzing Support Tickets. Classify the support tockets of the provided support ticket text into the following categories:\n",
        "    - Technical Issues\n",
        "    - Data Recovery\n",
        "    - Hardware Issues\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "Iatj3G2miXhn"
      },
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data_1['llama_response'] = data_1['support_ticket_text'].apply(lambda x: generate_llama_response(instruction_1,x))"
      ],
      "metadata": {
        "id": "iQN93JudirNb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2280ef4f-9024-4fcb-b2e3-4e8feb418d53"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      45.94 ms /    78 runs   (    0.59 ms per token,  1698.01 tokens per second)\n",
            "llama_print_timings: prompt eval time =     286.13 ms /   125 tokens (    2.29 ms per token,   436.87 tokens per second)\n",
            "llama_print_timings:        eval time =    1294.01 ms /    77 runs   (   16.81 ms per token,    59.51 tokens per second)\n",
            "llama_print_timings:       total time =    1838.31 ms /   202 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      17.20 ms /    29 runs   (    0.59 ms per token,  1686.34 tokens per second)\n",
            "llama_print_timings: prompt eval time =     178.90 ms /   123 tokens (    1.45 ms per token,   687.55 tokens per second)\n",
            "llama_print_timings:        eval time =     470.77 ms /    28 runs   (   16.81 ms per token,    59.48 tokens per second)\n",
            "llama_print_timings:       total time =     741.86 ms /   151 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      36.45 ms /    61 runs   (    0.60 ms per token,  1673.75 tokens per second)\n",
            "llama_print_timings: prompt eval time =     178.31 ms /   119 tokens (    1.50 ms per token,   667.37 tokens per second)\n",
            "llama_print_timings:        eval time =    1008.34 ms /    60 runs   (   16.81 ms per token,    59.50 tokens per second)\n",
            "llama_print_timings:       total time =    1381.54 ms /   179 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      74.16 ms /   130 runs   (    0.57 ms per token,  1753.06 tokens per second)\n",
            "llama_print_timings: prompt eval time =     179.81 ms /   127 tokens (    1.42 ms per token,   706.28 tokens per second)\n",
            "llama_print_timings:        eval time =    2177.04 ms /   129 runs   (   16.88 ms per token,    59.25 tokens per second)\n",
            "llama_print_timings:       total time =    2791.91 ms /   256 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      92.71 ms /   160 runs   (    0.58 ms per token,  1725.81 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.22 ms /   102 tokens (    1.72 ms per token,   582.14 tokens per second)\n",
            "llama_print_timings:        eval time =    2668.51 ms /   159 runs   (   16.78 ms per token,    59.58 tokens per second)\n",
            "llama_print_timings:       total time =    3381.24 ms /   261 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =     167.55 ms /   294 runs   (    0.57 ms per token,  1754.70 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.13 ms /   102 tokens (    1.72 ms per token,   582.41 tokens per second)\n",
            "llama_print_timings:        eval time =    4931.44 ms /   293 runs   (   16.83 ms per token,    59.41 tokens per second)\n",
            "llama_print_timings:       total time =    6107.28 ms /   395 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      72.19 ms /   120 runs   (    0.60 ms per token,  1662.35 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.94 ms /   105 tokens (    1.68 ms per token,   596.79 tokens per second)\n",
            "llama_print_timings:        eval time =    1983.59 ms /   119 runs   (   16.67 ms per token,    59.99 tokens per second)\n",
            "llama_print_timings:       total time =    2553.92 ms /   224 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      14.11 ms /    24 runs   (    0.59 ms per token,  1701.52 tokens per second)\n",
            "llama_print_timings: prompt eval time =     176.06 ms /   106 tokens (    1.66 ms per token,   602.08 tokens per second)\n",
            "llama_print_timings:        eval time =     383.04 ms /    23 runs   (   16.65 ms per token,    60.05 tokens per second)\n",
            "llama_print_timings:       total time =     639.55 ms /   129 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      50.33 ms /    88 runs   (    0.57 ms per token,  1748.56 tokens per second)\n",
            "llama_print_timings: prompt eval time =     174.85 ms /   100 tokens (    1.75 ms per token,   571.91 tokens per second)\n",
            "llama_print_timings:        eval time =    1453.71 ms /    87 runs   (   16.71 ms per token,    59.85 tokens per second)\n",
            "llama_print_timings:       total time =    1920.20 ms /   187 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      14.11 ms /    24 runs   (    0.59 ms per token,  1700.44 tokens per second)\n",
            "llama_print_timings: prompt eval time =     176.50 ms /   109 tokens (    1.62 ms per token,   617.57 tokens per second)\n",
            "llama_print_timings:        eval time =     382.88 ms /    23 runs   (   16.65 ms per token,    60.07 tokens per second)\n",
            "llama_print_timings:       total time =     639.72 ms /   132 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      19.91 ms /    34 runs   (    0.59 ms per token,  1707.94 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.16 ms /   102 tokens (    1.72 ms per token,   582.34 tokens per second)\n",
            "llama_print_timings:        eval time =     551.97 ms /    33 runs   (   16.73 ms per token,    59.79 tokens per second)\n",
            "llama_print_timings:       total time =     840.22 ms /   135 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      77.61 ms /   141 runs   (    0.55 ms per token,  1816.78 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.94 ms /   106 tokens (    1.66 ms per token,   602.49 tokens per second)\n",
            "llama_print_timings:        eval time =    2344.11 ms /   140 runs   (   16.74 ms per token,    59.72 tokens per second)\n",
            "llama_print_timings:       total time =    2978.81 ms /   246 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      90.93 ms /   158 runs   (    0.58 ms per token,  1737.52 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.88 ms /   106 tokens (    1.66 ms per token,   602.68 tokens per second)\n",
            "llama_print_timings:        eval time =    2625.05 ms /   157 runs   (   16.72 ms per token,    59.81 tokens per second)\n",
            "llama_print_timings:       total time =    3320.17 ms /   263 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      26.92 ms /    47 runs   (    0.57 ms per token,  1745.78 tokens per second)\n",
            "llama_print_timings: prompt eval time =     174.38 ms /    98 tokens (    1.78 ms per token,   561.98 tokens per second)\n",
            "llama_print_timings:        eval time =     761.78 ms /    46 runs   (   16.56 ms per token,    60.38 tokens per second)\n",
            "llama_print_timings:       total time =    1085.94 ms /   144 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      62.42 ms /   105 runs   (    0.59 ms per token,  1682.21 tokens per second)\n",
            "llama_print_timings: prompt eval time =     175.52 ms /   103 tokens (    1.70 ms per token,   586.84 tokens per second)\n",
            "llama_print_timings:        eval time =    1735.68 ms /   104 runs   (   16.69 ms per token,    59.92 tokens per second)\n",
            "llama_print_timings:       total time =    2253.65 ms /   207 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =     145.34 ms /   255 runs   (    0.57 ms per token,  1754.54 tokens per second)\n",
            "llama_print_timings: prompt eval time =     173.59 ms /    94 tokens (    1.85 ms per token,   541.52 tokens per second)\n",
            "llama_print_timings:        eval time =    4260.09 ms /   254 runs   (   16.77 ms per token,    59.62 tokens per second)\n",
            "llama_print_timings:       total time =    5283.08 ms /   348 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      96.20 ms /   175 runs   (    0.55 ms per token,  1819.13 tokens per second)\n",
            "llama_print_timings: prompt eval time =     174.66 ms /    99 tokens (    1.76 ms per token,   566.82 tokens per second)\n",
            "llama_print_timings:        eval time =    2909.73 ms /   174 runs   (   16.72 ms per token,    59.80 tokens per second)\n",
            "llama_print_timings:       total time =    3659.18 ms /   273 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      29.96 ms /    53 runs   (    0.57 ms per token,  1769.14 tokens per second)\n",
            "llama_print_timings: prompt eval time =     173.81 ms /    94 tokens (    1.85 ms per token,   540.81 tokens per second)\n",
            "llama_print_timings:        eval time =     866.84 ms /    52 runs   (   16.67 ms per token,    59.99 tokens per second)\n",
            "llama_print_timings:       total time =    1219.36 ms /   146 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      86.31 ms /   148 runs   (    0.58 ms per token,  1714.85 tokens per second)\n",
            "llama_print_timings: prompt eval time =     174.05 ms /    95 tokens (    1.83 ms per token,   545.83 tokens per second)\n",
            "llama_print_timings:        eval time =    2461.24 ms /   147 runs   (   16.74 ms per token,    59.73 tokens per second)\n",
            "llama_print_timings:       total time =    3137.17 ms /   242 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      48.08 ms /    84 runs   (    0.57 ms per token,  1747.09 tokens per second)\n",
            "llama_print_timings: prompt eval time =     204.35 ms /   251 tokens (    0.81 ms per token,  1228.30 tokens per second)\n",
            "llama_print_timings:        eval time =    1400.35 ms /    83 runs   (   16.87 ms per token,    59.27 tokens per second)\n",
            "llama_print_timings:       total time =    1871.97 ms /   334 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =     116.49 ms /   211 runs   (    0.55 ms per token,  1811.38 tokens per second)\n",
            "llama_print_timings: prompt eval time =     243.93 ms /   364 tokens (    0.67 ms per token,  1492.23 tokens per second)\n",
            "llama_print_timings:        eval time =    3600.25 ms /   210 runs   (   17.14 ms per token,    58.33 tokens per second)\n",
            "llama_print_timings:       total time =    4548.46 ms /   574 tokens\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_1.head()\n"
      ],
      "metadata": {
        "id": "_Yqwune_i6Ws",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "7cd34372-9000-4f8e-8123-079478f2d7bb"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \n",
              "0   Based on the provided support ticket text, I ...  \n",
              "1   Based on the provided support ticket text, th...  \n",
              "2   Based on the provided support ticket text, I ...  \n",
              "3   Based on the provided support ticket text, I ...  \n",
              "4   Sure! Based on the given support ticket text,...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0bd504d5-1a8a-4f11-b3a9-36316df96301\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Based on the provided support ticket text, I ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Based on the provided support ticket text, th...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Based on the provided support ticket text, I ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Based on the provided support ticket text, I ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Based on the given support ticket text,...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0bd504d5-1a8a-4f11-b3a9-36316df96301')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-0bd504d5-1a8a-4f11-b3a9-36316df96301 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-0bd504d5-1a8a-4f11-b3a9-36316df96301');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-5bcf1805-0ceb-4e3f-901a-92b7f5dce7fe\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-5bcf1805-0ceb-4e3f-901a-92b7f5dce7fe')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-5bcf1805-0ceb-4e3f-901a-92b7f5dce7fe button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_1",
              "summary": "{\n  \"name\": \"data_1\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" Based on the provided support ticket text, I would classify the support ticket as follows:\\n\\n     Technical Issues: This support ticket is related to a technical issue with the customer's internet connection, specifically frequent disconnections and a significant slowdown in speed.\\n    \\n     Therefore, the category for this support ticket would be \\\"Technical Issues\\\".\",\n          \" Based on the content of the given support ticket text, I would classify it as follows:\\n\\n* Technical Issues: 60% (the external hard drive not being recognized)\\n* Data Recovery: 40% (vital data recovery)\\n\\nHere is a breakdown of the classification by category:\\n\\n* Technical Issues:\\n\\t+ The external hard drive not being recognized is a technical issue related to hardware or software compatibility, driver installation, or other technical problems.\\n* Data Recovery:\\n\\t+ Vital data recovery is a data recovery issue related to retrieving lost or corrupted files from a storage device, which is a common request in technical support.\",\n          \" Sure, I can help you with that! Based on the provided support ticket text, I would classify it as follows:\\n\\n* Technical Issues: 80%\\n\\t+ The user is experiencing inconsistent Wi-Fi connectivity, which suggests a technical issue with the network or router settings.\\n* Data Recovery: 10%\\n\\t+ The user mentions that their work is affected by the inconsistent Wi-Fi connectivity, which could potentially result in data loss or corruption if not resolved.\\n* Hardware Issues: 10%\\n\\t+ The user mentions that they have tried restarting their router, which suggests a possible hardware issue with the router or network device.\\n\\nPlease note that these percentages are approximate and may vary depending on the specific context and details of the support ticket.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def extract_category(model_response):\n",
        "    if 'technical issues' in model_response.lower():\n",
        "        return 'Technical issues'\n",
        "    elif 'hardware issues' in model_response.lower():\n",
        "        return 'Hardware issues'\n",
        "    elif 'data recovery' in model_response.lower():\n",
        "        return 'Data recovery'"
      ],
      "metadata": {
        "id": "SIcBrC4LjA5L"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data_1['Category'] = data_1['llama_response'].apply(extract_category)\n",
        "data_1['Category'].head()"
      ],
      "metadata": {
        "id": "QJ87PpFFjDJT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1d10216f-c241-4d24-8bd2-dff52996fd85"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    Technical issues\n",
              "1     Hardware issues\n",
              "2       Data recovery\n",
              "3    Technical issues\n",
              "4    Technical issues\n",
              "Name: Category, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data_1 = data_1.drop(['llama_response'], axis=1)\n",
        "final_data_1.head()"
      ],
      "metadata": {
        "id": "eMxgcOuPjO8E",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "865c59b6-e10e-4b40-8b93-4a459feedfaf"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "           Category  \n",
              "0  Technical issues  \n",
              "1   Hardware issues  \n",
              "2     Data recovery  \n",
              "3  Technical issues  \n",
              "4  Technical issues  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-cda784dc-61a5-46eb-b87c-ea52232bbc84\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>Category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Technical issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Hardware issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Data recovery</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Technical issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Technical issues</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cda784dc-61a5-46eb-b87c-ea52232bbc84')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-cda784dc-61a5-46eb-b87c-ea52232bbc84 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-cda784dc-61a5-46eb-b87c-ea52232bbc84');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-4d2f0fa3-d4c1-407d-bcc4-86e50907b263\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-4d2f0fa3-d4c1-407d-bcc4-86e50907b263')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-4d2f0fa3-d4c1-407d-bcc4-86e50907b263 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_1",
              "summary": "{\n  \"name\": \"final_data_1\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical issues\",\n          \"Hardware issues\",\n          \"Data recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Task 2: Ticket Categorization and Returning Structured Output**"
      ],
      "metadata": {
        "id": "mukxSgHYkqqt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# create a copy of the data\n",
        "data_2 = data.copy()"
      ],
      "metadata": {
        "id": "VS4rU0QbjZHA"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the prompt to get the desired output\n",
        "instruction_2 =  \"\"\"\n",
        "You are an AI analyzing Support Tickets. Classify the support tockets of the provided support ticket text into the following categories:\n",
        "    - Technical Issues\n",
        "    - Data Recovery\n",
        "    - Hardware Issues\n",
        "\n",
        "    Format the output as a JSON object with key-value pair as shown below:\n",
        "    {\n",
        "      \"Category\": \"your_category_prediction\"\n",
        "    }\n",
        "\n",
        "     Only return the JSON, do NOT return any other text or information.\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "_WPMLKK8k0GF"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data_2['llama_response'] = data_2['support_ticket_text'].apply(lambda x: generate_llama_response(instruction_2,x))"
      ],
      "metadata": {
        "id": "b-BioKYAk7CR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "81a9a7e2-3be7-4935-d95b-a62b852fb39d"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.78 ms /    17 runs   (    0.58 ms per token,  1738.42 tokens per second)\n",
            "llama_print_timings: prompt eval time =     216.26 ms /   179 tokens (    1.21 ms per token,   827.69 tokens per second)\n",
            "llama_print_timings:        eval time =     269.15 ms /    16 runs   (   16.82 ms per token,    59.45 tokens per second)\n",
            "llama_print_timings:       total time =     544.13 ms /   195 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.84 ms /    17 runs   (    0.58 ms per token,  1727.47 tokens per second)\n",
            "llama_print_timings: prompt eval time =     199.87 ms /   178 tokens (    1.12 ms per token,   890.59 tokens per second)\n",
            "llama_print_timings:        eval time =     268.66 ms /    16 runs   (   16.79 ms per token,    59.55 tokens per second)\n",
            "llama_print_timings:       total time =     523.08 ms /   194 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.23 ms /    15 runs   (    0.62 ms per token,  1625.84 tokens per second)\n",
            "llama_print_timings: prompt eval time =     200.35 ms /   174 tokens (    1.15 ms per token,   868.50 tokens per second)\n",
            "llama_print_timings:        eval time =     236.18 ms /    14 runs   (   16.87 ms per token,    59.28 tokens per second)\n",
            "llama_print_timings:       total time =     488.15 ms /   188 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.71 ms /    17 runs   (    0.57 ms per token,  1750.05 tokens per second)\n",
            "llama_print_timings: prompt eval time =     199.43 ms /   182 tokens (    1.10 ms per token,   912.61 tokens per second)\n",
            "llama_print_timings:        eval time =     270.53 ms /    16 runs   (   16.91 ms per token,    59.14 tokens per second)\n",
            "llama_print_timings:       total time =     527.30 ms /   198 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.86 ms /    17 runs   (    0.58 ms per token,  1723.79 tokens per second)\n",
            "llama_print_timings: prompt eval time =     201.72 ms /   157 tokens (    1.28 ms per token,   778.31 tokens per second)\n",
            "llama_print_timings:        eval time =     268.46 ms /    16 runs   (   16.78 ms per token,    59.60 tokens per second)\n",
            "llama_print_timings:       total time =     525.60 ms /   173 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.41 ms /    17 runs   (    0.55 ms per token,  1807.16 tokens per second)\n",
            "llama_print_timings: prompt eval time =     201.61 ms /   157 tokens (    1.28 ms per token,   778.74 tokens per second)\n",
            "llama_print_timings:        eval time =     268.58 ms /    16 runs   (   16.79 ms per token,    59.57 tokens per second)\n",
            "llama_print_timings:       total time =     525.24 ms /   173 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.91 ms /    17 runs   (    0.58 ms per token,  1716.13 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.44 ms /   160 tokens (    1.27 ms per token,   790.38 tokens per second)\n",
            "llama_print_timings:        eval time =     269.43 ms /    16 runs   (   16.84 ms per token,    59.38 tokens per second)\n",
            "llama_print_timings:       total time =     528.58 ms /   176 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      10.25 ms /    17 runs   (    0.60 ms per token,  1659.35 tokens per second)\n",
            "llama_print_timings: prompt eval time =     201.72 ms /   161 tokens (    1.25 ms per token,   798.13 tokens per second)\n",
            "llama_print_timings:        eval time =     270.17 ms /    16 runs   (   16.89 ms per token,    59.22 tokens per second)\n",
            "llama_print_timings:       total time =     529.23 ms /   177 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       8.61 ms /    15 runs   (    0.57 ms per token,  1742.77 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.00 ms /   155 tokens (    1.30 ms per token,   767.34 tokens per second)\n",
            "llama_print_timings:        eval time =     235.60 ms /    14 runs   (   16.83 ms per token,    59.42 tokens per second)\n",
            "llama_print_timings:       total time =     488.68 ms /   169 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.93 ms /    17 runs   (    0.58 ms per token,  1711.29 tokens per second)\n",
            "llama_print_timings: prompt eval time =     201.46 ms /   164 tokens (    1.23 ms per token,   814.08 tokens per second)\n",
            "llama_print_timings:        eval time =     269.07 ms /    16 runs   (   16.82 ms per token,    59.46 tokens per second)\n",
            "llama_print_timings:       total time =     528.25 ms /   180 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      20.43 ms /    33 runs   (    0.62 ms per token,  1615.11 tokens per second)\n",
            "llama_print_timings: prompt eval time =     200.97 ms /   157 tokens (    1.28 ms per token,   781.23 tokens per second)\n",
            "llama_print_timings:        eval time =     538.02 ms /    32 runs   (   16.81 ms per token,    59.48 tokens per second)\n",
            "llama_print_timings:       total time =     847.80 ms /   189 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      10.00 ms /    17 runs   (    0.59 ms per token,  1700.85 tokens per second)\n",
            "llama_print_timings: prompt eval time =     201.32 ms /   161 tokens (    1.25 ms per token,   799.71 tokens per second)\n",
            "llama_print_timings:        eval time =     270.49 ms /    16 runs   (   16.91 ms per token,    59.15 tokens per second)\n",
            "llama_print_timings:       total time =     524.97 ms /   177 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       8.49 ms /    15 runs   (    0.57 ms per token,  1766.99 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.08 ms /   161 tokens (    1.26 ms per token,   796.72 tokens per second)\n",
            "llama_print_timings:        eval time =     235.78 ms /    14 runs   (   16.84 ms per token,    59.38 tokens per second)\n",
            "llama_print_timings:       total time =     485.86 ms /   175 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       8.38 ms /    15 runs   (    0.56 ms per token,  1789.12 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.04 ms /   153 tokens (    1.32 ms per token,   757.26 tokens per second)\n",
            "llama_print_timings:        eval time =     234.80 ms /    14 runs   (   16.77 ms per token,    59.63 tokens per second)\n",
            "llama_print_timings:       total time =     485.30 ms /   167 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      10.27 ms /    17 runs   (    0.60 ms per token,  1655.95 tokens per second)\n",
            "llama_print_timings: prompt eval time =     200.64 ms /   158 tokens (    1.27 ms per token,   787.49 tokens per second)\n",
            "llama_print_timings:        eval time =     270.19 ms /    16 runs   (   16.89 ms per token,    59.22 tokens per second)\n",
            "llama_print_timings:       total time =     530.51 ms /   174 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      10.35 ms /    17 runs   (    0.61 ms per token,  1641.88 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.71 ms /   149 tokens (    1.36 ms per token,   735.03 tokens per second)\n",
            "llama_print_timings:        eval time =     269.05 ms /    16 runs   (   16.82 ms per token,    59.47 tokens per second)\n",
            "llama_print_timings:       total time =     529.87 ms /   165 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      10.16 ms /    17 runs   (    0.60 ms per token,  1673.56 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.38 ms /   154 tokens (    1.31 ms per token,   760.95 tokens per second)\n",
            "llama_print_timings:        eval time =     271.24 ms /    16 runs   (   16.95 ms per token,    58.99 tokens per second)\n",
            "llama_print_timings:       total time =     534.36 ms /   170 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       8.95 ms /    15 runs   (    0.60 ms per token,  1676.35 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.86 ms /   149 tokens (    1.36 ms per token,   734.49 tokens per second)\n",
            "llama_print_timings:        eval time =     236.46 ms /    14 runs   (   16.89 ms per token,    59.21 tokens per second)\n",
            "llama_print_timings:       total time =     492.56 ms /   163 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       8.62 ms /    15 runs   (    0.57 ms per token,  1740.75 tokens per second)\n",
            "llama_print_timings: prompt eval time =     202.71 ms /   150 tokens (    1.35 ms per token,   739.98 tokens per second)\n",
            "llama_print_timings:        eval time =     233.47 ms /    14 runs   (   16.68 ms per token,    59.97 tokens per second)\n",
            "llama_print_timings:       total time =     484.12 ms /   164 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.47 ms /    17 runs   (    0.56 ms per token,  1794.57 tokens per second)\n",
            "llama_print_timings: prompt eval time =     236.49 ms /   306 tokens (    0.77 ms per token,  1293.92 tokens per second)\n",
            "llama_print_timings:        eval time =     271.49 ms /    16 runs   (   16.97 ms per token,    58.93 tokens per second)\n",
            "llama_print_timings:       total time =     562.01 ms /   322 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =       9.65 ms /    17 runs   (    0.57 ms per token,  1762.39 tokens per second)\n",
            "llama_print_timings: prompt eval time =     275.60 ms /   419 tokens (    0.66 ms per token,  1520.31 tokens per second)\n",
            "llama_print_timings:        eval time =     274.11 ms /    16 runs   (   17.13 ms per token,    58.37 tokens per second)\n",
            "llama_print_timings:       total time =     606.00 ms /   435 tokens\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_2.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "VQNMflgXXZkl",
        "outputId": "4e6df574-6534-4b34-edc5-f44dcc273b0d"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                     llama_response  \n",
              "0    {\\n      \"Category\": \"Technical Issues\"\\n    }  \n",
              "1   {\\n       \"Category\": \"Hardware Issues\"\\n     }  \n",
              "2     {\\n       \"Category\": \"Data Recovery\"\\n     }  \n",
              "3    {\\n      \"Category\": \"Technical Issues\"\\n    }  \n",
              "4     {\\n      \"Category\": \"Hardware Issues\"\\n    }  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-089dd4b2-f544-4f50-8d16-0fd6d22dd76e\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\"\\n    }</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n       \"Category\": \"Hardware Issues\"\\n     }</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n       \"Category\": \"Data Recovery\"\\n     }</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\"\\n    }</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\"\\n    }</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-089dd4b2-f544-4f50-8d16-0fd6d22dd76e')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-089dd4b2-f544-4f50-8d16-0fd6d22dd76e button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-089dd4b2-f544-4f50-8d16-0fd6d22dd76e');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-5b9871df-1e1f-4f3c-a7e3-658bbdfd1855\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-5b9871df-1e1f-4f3c-a7e3-658bbdfd1855')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-5b9871df-1e1f-4f3c-a7e3-658bbdfd1855 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_2",
              "summary": "{\n  \"name\": \"data_2\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \" {\\n       \\\"Category\\\": \\\"Hardware Issues\\\"\\n     }\",\n          \" Sure! Based on the content of your support ticket, I classify it as follows:\\n\\n{\\n\\\"Category\\\": \\\"Data Recovery\\\"\\n}\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\"\\n    }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data_2 = data_2.drop(['llama_response'], axis=1)\n",
        "final_data_2.head()"
      ],
      "metadata": {
        "id": "DYc4H8e_lGaH",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "4938d8c5-d37d-4e15-fa99-9006157f0e19"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text\n",
              "0      ST2023-006  My internet connection has significantly slowe...\n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...\n",
              "2      ST2023-008  I've accidentally deleted essential work docum...\n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...\n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve..."
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f76d6fe4-c183-44e1-b699-88eb1d617c52\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f76d6fe4-c183-44e1-b699-88eb1d617c52')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-f76d6fe4-c183-44e1-b699-88eb1d617c52 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-f76d6fe4-c183-44e1-b699-88eb1d617c52');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-81650419-11df-4ce9-86fc-31c457120e48\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-81650419-11df-4ce9-86fc-31c457120e48')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-81650419-11df-4ce9-86fc-31c457120e48 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_2",
              "summary": "{\n  \"name\": \"final_data_2\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# defining a function to parse the JSON output from the model\n",
        "def extract_json_data(json_str):\n",
        "    try:\n",
        "        # Find the indices of the opening and closing curly braces\n",
        "        json_start = json_str.find('{')\n",
        "        json_end = json_str.rfind('}')\n",
        "\n",
        "        if json_start != -1 and json_end != -1:\n",
        "            extracted_category = json_str[json_start:json_end + 1]  # Extract the JSON object\n",
        "            data_dict = json.loads(extracted_category)\n",
        "            return data_dict\n",
        "        else:\n",
        "            print(f\"Warning: JSON object not found in response: {json_str}\")\n",
        "            return {}\n",
        "    except json.JSONDecodeError as e:\n",
        "        print(f\"Error parsing JSON: {e}\")\n",
        "        return {}"
      ],
      "metadata": {
        "id": "NZzciz0flJ8l"
      },
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data_2['llama_response_parsed'] = data_2['llama_response'].apply(extract_json_data)\n",
        "data_2['llama_response_parsed'].head()"
      ],
      "metadata": {
        "id": "lmHx78hwlKjH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5d1b91f5-df7e-4094-86a6-55d3a0a83bea"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    {'Category': 'Technical Issues'}\n",
              "1     {'Category': 'Hardware Issues'}\n",
              "2       {'Category': 'Data Recovery'}\n",
              "3    {'Category': 'Technical Issues'}\n",
              "4     {'Category': 'Hardware Issues'}\n",
              "Name: llama_response_parsed, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "llama_response_parsed_df_2 = pd.json_normalize(data_2['llama_response_parsed'])\n",
        "llama_response_parsed_df_2.head()"
      ],
      "metadata": {
        "id": "Z5WwtJsLlmOV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "d9cfc280-63d9-4748-e964-4dd4a4c85867"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Category\n",
              "0  Technical Issues\n",
              "1   Hardware Issues\n",
              "2     Data Recovery\n",
              "3  Technical Issues\n",
              "4   Hardware Issues"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0a2c595e-bedb-413c-8ba5-e3e9fbd6eeec\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Technical Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Hardware Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Data Recovery</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Technical Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Hardware Issues</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0a2c595e-bedb-413c-8ba5-e3e9fbd6eeec')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-0a2c595e-bedb-413c-8ba5-e3e9fbd6eeec button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-0a2c595e-bedb-413c-8ba5-e3e9fbd6eeec');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-d69241ce-3d4e-47b7-b2cb-645e607b0e6d\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-d69241ce-3d4e-47b7-b2cb-645e607b0e6d')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-d69241ce-3d4e-47b7-b2cb-645e607b0e6d button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "llama_response_parsed_df_2",
              "summary": "{\n  \"name\": \"llama_response_parsed_df_2\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_with_parsed_model_output_2 = pd.concat([data_2, llama_response_parsed_df_2], axis=1)\n",
        "data_with_parsed_model_output_2.head()"
      ],
      "metadata": {
        "id": "35aY9Rr_lwJO",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "154d6b50-b1fd-4600-a169-2e6c9f3b7728"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                     llama_response  \\\n",
              "0    {\\n      \"Category\": \"Technical Issues\"\\n    }   \n",
              "1   {\\n       \"Category\": \"Hardware Issues\"\\n     }   \n",
              "2     {\\n       \"Category\": \"Data Recovery\"\\n     }   \n",
              "3    {\\n      \"Category\": \"Technical Issues\"\\n    }   \n",
              "4     {\\n      \"Category\": \"Hardware Issues\"\\n    }   \n",
              "\n",
              "              llama_response_parsed          Category  \n",
              "0  {'Category': 'Technical Issues'}  Technical Issues  \n",
              "1   {'Category': 'Hardware Issues'}   Hardware Issues  \n",
              "2     {'Category': 'Data Recovery'}     Data Recovery  \n",
              "3  {'Category': 'Technical Issues'}  Technical Issues  \n",
              "4   {'Category': 'Hardware Issues'}   Hardware Issues  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7812c4e7-0ba3-4b59-a455-d69739bf71d9\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "      <th>llama_response_parsed</th>\n",
              "      <th>Category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\"\\n    }</td>\n",
              "      <td>{'Category': 'Technical Issues'}</td>\n",
              "      <td>Technical Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n       \"Category\": \"Hardware Issues\"\\n     }</td>\n",
              "      <td>{'Category': 'Hardware Issues'}</td>\n",
              "      <td>Hardware Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n       \"Category\": \"Data Recovery\"\\n     }</td>\n",
              "      <td>{'Category': 'Data Recovery'}</td>\n",
              "      <td>Data Recovery</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\"\\n    }</td>\n",
              "      <td>{'Category': 'Technical Issues'}</td>\n",
              "      <td>Technical Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\"\\n    }</td>\n",
              "      <td>{'Category': 'Hardware Issues'}</td>\n",
              "      <td>Hardware Issues</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7812c4e7-0ba3-4b59-a455-d69739bf71d9')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-7812c4e7-0ba3-4b59-a455-d69739bf71d9 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-7812c4e7-0ba3-4b59-a455-d69739bf71d9');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-4bcc862f-793f-4bae-b54e-ededab02e530\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-4bcc862f-793f-4bae-b54e-ededab02e530')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-4bcc862f-793f-4bae-b54e-ededab02e530 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_with_parsed_model_output_2",
              "summary": "{\n  \"name\": \"data_with_parsed_model_output_2\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \" {\\n       \\\"Category\\\": \\\"Hardware Issues\\\"\\n     }\",\n          \" Sure! Based on the content of your support ticket, I classify it as follows:\\n\\n{\\n\\\"Category\\\": \\\"Data Recovery\\\"\\n}\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\"\\n    }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response_parsed\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data_2 = data_with_parsed_model_output_2.drop(['llama_response','llama_response_parsed'], axis=1)\n",
        "final_data_2.head()"
      ],
      "metadata": {
        "id": "XRgxHjgJl6os",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "d3363b53-8bdb-4b76-f91b-7fc4de24a83e"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "           Category  \n",
              "0  Technical Issues  \n",
              "1   Hardware Issues  \n",
              "2     Data Recovery  \n",
              "3  Technical Issues  \n",
              "4   Hardware Issues  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b4e3d3cd-a969-4d17-aa3f-0749e8a5967f\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>Category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Technical Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Data Recovery</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Technical Issues</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b4e3d3cd-a969-4d17-aa3f-0749e8a5967f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-b4e3d3cd-a969-4d17-aa3f-0749e8a5967f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-b4e3d3cd-a969-4d17-aa3f-0749e8a5967f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-b76e5ca0-f714-42c7-841d-e1f07b16e63c\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-b76e5ca0-f714-42c7-841d-e1f07b16e63c')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-b76e5ca0-f714-42c7-841d-e1f07b16e63c button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_2",
              "summary": "{\n  \"name\": \"final_data_2\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Task 3: Ticket Categorization, Creating Tags, and Returning Structured Output**"
      ],
      "metadata": {
        "id": "bkDWej0_mifs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# create a copy of the data\n",
        "data_3 = data.copy()"
      ],
      "metadata": {
        "id": "DYLkkZE6mpwN"
      },
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the prompt to get the desired output\n",
        "instruction_3 = \"\"\"\n",
        "    You are an AI analyzing Support Tickets. Classify the support tockets of the provided support ticket text into the following categories:\n",
        "    - Technical Issues\n",
        "    - Data Recovery\n",
        "    - Hardware Issues\n",
        "\n",
        "    Tags should reflect specific problem areas, technologies involved, add support ticket category and actions required.\n",
        "\n",
        "    Provide the output in a JSON format with the following keys:\n",
        "    {\n",
        "      \"Category\": \"your_category_prediction\",\n",
        "      \"Tags\": \"your_tags_prediction\"\n",
        "    }\n",
        "\n",
        "      Tags should be list item.\n",
        "      Only return the JSON, do NOT return any other text or information.\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "VZWajMFhnIQH"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "data_3['llama_response'] = data_3['support_ticket_text'].apply(lambda x: generate_llama_response(instruction_3,x))"
      ],
      "metadata": {
        "id": "NRes28rHgGEv",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "119a040f-22d0-4b59-9ab6-ee6d73dcc2ed"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      28.98 ms /    51 runs   (    0.57 ms per token,  1760.02 tokens per second)\n",
            "llama_print_timings: prompt eval time =     218.30 ms /   220 tokens (    0.99 ms per token,  1007.80 tokens per second)\n",
            "llama_print_timings:        eval time =     843.17 ms /    50 runs   (   16.86 ms per token,    59.30 tokens per second)\n",
            "llama_print_timings:       total time =    1226.25 ms /   270 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.23 ms /    48 runs   (    0.57 ms per token,  1762.83 tokens per second)\n",
            "llama_print_timings: prompt eval time =     199.83 ms /   219 tokens (    0.91 ms per token,  1095.95 tokens per second)\n",
            "llama_print_timings:        eval time =     790.25 ms /    47 runs   (   16.81 ms per token,    59.48 tokens per second)\n",
            "llama_print_timings:       total time =    1146.41 ms /   266 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      22.54 ms /    40 runs   (    0.56 ms per token,  1774.54 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.99 ms /   215 tokens (    0.93 ms per token,  1080.45 tokens per second)\n",
            "llama_print_timings:        eval time =     653.47 ms /    39 runs   (   16.76 ms per token,    59.68 tokens per second)\n",
            "llama_print_timings:       total time =     979.68 ms /   254 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      21.71 ms /    40 runs   (    0.54 ms per token,  1842.81 tokens per second)\n",
            "llama_print_timings: prompt eval time =     200.33 ms /   223 tokens (    0.90 ms per token,  1113.18 tokens per second)\n",
            "llama_print_timings:        eval time =     657.14 ms /    39 runs   (   16.85 ms per token,    59.35 tokens per second)\n",
            "llama_print_timings:       total time =     981.68 ms /   262 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      77.33 ms /   136 runs   (    0.57 ms per token,  1758.79 tokens per second)\n",
            "llama_print_timings: prompt eval time =     199.55 ms /   198 tokens (    1.01 ms per token,   992.26 tokens per second)\n",
            "llama_print_timings:        eval time =    2271.52 ms /   135 runs   (   16.83 ms per token,    59.43 tokens per second)\n",
            "llama_print_timings:       total time =    2911.72 ms /   333 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      35.07 ms /    60 runs   (    0.58 ms per token,  1711.01 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.38 ms /   198 tokens (    1.00 ms per token,   998.11 tokens per second)\n",
            "llama_print_timings:        eval time =     996.35 ms /    59 runs   (   16.89 ms per token,    59.22 tokens per second)\n",
            "llama_print_timings:       total time =    1396.18 ms /   257 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      30.52 ms /    54 runs   (    0.57 ms per token,  1769.62 tokens per second)\n",
            "llama_print_timings: prompt eval time =     197.98 ms /   201 tokens (    0.98 ms per token,  1015.25 tokens per second)\n",
            "llama_print_timings:        eval time =     889.46 ms /    53 runs   (   16.78 ms per token,    59.59 tokens per second)\n",
            "llama_print_timings:       total time =    1256.61 ms /   254 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      22.68 ms /    39 runs   (    0.58 ms per token,  1719.96 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.22 ms /   202 tokens (    0.98 ms per token,  1019.04 tokens per second)\n",
            "llama_print_timings:        eval time =     636.56 ms /    38 runs   (   16.75 ms per token,    59.70 tokens per second)\n",
            "llama_print_timings:       total time =     963.63 ms /   240 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      19.82 ms /    35 runs   (    0.57 ms per token,  1765.98 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.09 ms /   196 tokens (    1.01 ms per token,   989.46 tokens per second)\n",
            "llama_print_timings:        eval time =     570.89 ms /    34 runs   (   16.79 ms per token,    59.56 tokens per second)\n",
            "llama_print_timings:       total time =     882.28 ms /   230 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.97 ms /    49 runs   (    0.57 ms per token,  1751.88 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.13 ms /   205 tokens (    0.97 ms per token,  1034.68 tokens per second)\n",
            "llama_print_timings:        eval time =     804.68 ms /    48 runs   (   16.76 ms per token,    59.65 tokens per second)\n",
            "llama_print_timings:       total time =    1159.49 ms /   253 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      38.30 ms /    68 runs   (    0.56 ms per token,  1775.55 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.43 ms /   198 tokens (    1.00 ms per token,   997.81 tokens per second)\n",
            "llama_print_timings:        eval time =    1125.96 ms /    67 runs   (   16.81 ms per token,    59.50 tokens per second)\n",
            "llama_print_timings:       total time =    1542.67 ms /   265 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      18.22 ms /    32 runs   (    0.57 ms per token,  1755.93 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.31 ms /   202 tokens (    0.98 ms per token,  1018.61 tokens per second)\n",
            "llama_print_timings:        eval time =     520.42 ms /    31 runs   (   16.79 ms per token,    59.57 tokens per second)\n",
            "llama_print_timings:       total time =     819.58 ms /   233 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      22.39 ms /    37 runs   (    0.61 ms per token,  1652.60 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.38 ms /   202 tokens (    0.98 ms per token,  1018.24 tokens per second)\n",
            "llama_print_timings:        eval time =     604.58 ms /    36 runs   (   16.79 ms per token,    59.55 tokens per second)\n",
            "llama_print_timings:       total time =     925.94 ms /   238 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      29.95 ms /    51 runs   (    0.59 ms per token,  1703.07 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.16 ms /   194 tokens (    1.02 ms per token,   979.00 tokens per second)\n",
            "llama_print_timings:        eval time =     841.07 ms /    50 runs   (   16.82 ms per token,    59.45 tokens per second)\n",
            "llama_print_timings:       total time =    1212.23 ms /   244 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.75 ms /    46 runs   (    0.60 ms per token,  1657.90 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.58 ms /   199 tokens (    1.00 ms per token,  1002.14 tokens per second)\n",
            "llama_print_timings:        eval time =     758.84 ms /    45 runs   (   16.86 ms per token,    59.30 tokens per second)\n",
            "llama_print_timings:       total time =    1110.59 ms /   244 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      18.96 ms /    33 runs   (    0.57 ms per token,  1740.41 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.74 ms /   190 tokens (    1.05 ms per token,   956.04 tokens per second)\n",
            "llama_print_timings:        eval time =     537.50 ms /    32 runs   (   16.80 ms per token,    59.53 tokens per second)\n",
            "llama_print_timings:       total time =     848.51 ms /   222 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      34.58 ms /    63 runs   (    0.55 ms per token,  1821.92 tokens per second)\n",
            "llama_print_timings: prompt eval time =     198.58 ms /   195 tokens (    1.02 ms per token,   982.00 tokens per second)\n",
            "llama_print_timings:        eval time =    1041.08 ms /    62 runs   (   16.79 ms per token,    59.55 tokens per second)\n",
            "llama_print_timings:       total time =    1448.55 ms /   257 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      79.04 ms /   141 runs   (    0.56 ms per token,  1784.00 tokens per second)\n",
            "llama_print_timings: prompt eval time =     197.75 ms /   190 tokens (    1.04 ms per token,   960.83 tokens per second)\n",
            "llama_print_timings:        eval time =    2353.47 ms /   140 runs   (   16.81 ms per token,    59.49 tokens per second)\n",
            "llama_print_timings:       total time =    3013.78 ms /   330 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      20.82 ms /    35 runs   (    0.59 ms per token,  1681.00 tokens per second)\n",
            "llama_print_timings: prompt eval time =     197.28 ms /   191 tokens (    1.03 ms per token,   968.14 tokens per second)\n",
            "llama_print_timings:        eval time =     571.05 ms /    34 runs   (   16.80 ms per token,    59.54 tokens per second)\n",
            "llama_print_timings:       total time =     882.37 ms /   225 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      26.26 ms /    46 runs   (    0.57 ms per token,  1751.78 tokens per second)\n",
            "llama_print_timings: prompt eval time =     241.53 ms /   347 tokens (    0.70 ms per token,  1436.66 tokens per second)\n",
            "llama_print_timings:        eval time =     762.74 ms /    45 runs   (   16.95 ms per token,    59.00 tokens per second)\n",
            "llama_print_timings:       total time =    1152.02 ms /   392 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      22.83 ms /    40 runs   (    0.57 ms per token,  1752.39 tokens per second)\n",
            "llama_print_timings: prompt eval time =     279.48 ms /   460 tokens (    0.61 ms per token,  1645.92 tokens per second)\n",
            "llama_print_timings:        eval time =     668.85 ms /    39 runs   (   17.15 ms per token,    58.31 tokens per second)\n",
            "llama_print_timings:       total time =    1079.22 ms /   499 tokens\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Write the code to check the first five rows of the data to confirm whether the new column has been added\n",
        "data_3.head()"
      ],
      "metadata": {
        "id": "vC75jBhpnYBp",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "401864ab-4431-44f5-c60d-531ea7cddf36"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \n",
              "0   Sure! Here is my analysis of your support tic...  \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...  \n",
              "2   {\\n      \"Category\": \"Data Recovery\",\\n      ...  \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...  \n",
              "4   Sure! Here's my analysis of your support tick...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-63611544-c30f-4e6c-9ee4-a0d1771fd189\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Sure! Here is my analysis of your support tic...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n      \"Category\": \"Data Recovery\",\\n      ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Here's my analysis of your support tick...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-63611544-c30f-4e6c-9ee4-a0d1771fd189')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-63611544-c30f-4e6c-9ee4-a0d1771fd189 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-63611544-c30f-4e6c-9ee4-a0d1771fd189');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-0b7c326f-4baa-4a8e-8370-956cdea75f14\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0b7c326f-4baa-4a8e-8370-956cdea75f14')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-0b7c326f-4baa-4a8e-8370-956cdea75f14 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_3",
              "summary": "{\n  \"name\": \"data_3\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" Sure! Here is my analysis of your support ticket text:\\n    \\n   {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connectivity\\\", \\\"Disconnections\\\", \\\"Speed\\\"]\\n   }\",\n          \" Sure, I can help you classify your support tickets! Based on the information you've given me, here is my prediction for your first support ticket:\\n\\n{\\n  \\\"Category\\\": \\\"Data Recovery\\\",\\n  \\\"Tags\\\": [\\\"USB Drive\\\", \\\"File Recovery\\\", \\\"Crucial Files\\\"]\\n}\\n\\nBased on the information in your support ticket, it seems like you have encountered a data loss issue and are seeking help to recover your crucial files from a formatted USB drive. This falls under the category of Data Recovery, and some relevant tags that could be applied to this ticket include \\\"USB Drive\\\" and \\\"File Recovery\\\".\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Drop\\\", \\\"Frequent\\\"]\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_3['llama_response_parsed'] = data_3['llama_response'].apply(extract_json_data)\n",
        "data_3['llama_response_parsed'].head()"
      ],
      "metadata": {
        "id": "DAvvphUdggEf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f97031a6-08db-449a-cf73-d76272628ce9"
      },
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    {'Category': 'Technical Issues', 'Tags': ['Int...\n",
              "1    {'Category': 'Hardware Issues', 'Tags': ['Lapt...\n",
              "2    {'Category': 'Data Recovery', 'Tags': ['Data L...\n",
              "3    {'Category': 'Technical Issues', 'Tags': ['Wi-...\n",
              "4    {'Category': 'Hardware Issues', 'Tags': ['Batt...\n",
              "Name: llama_response_parsed, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "llama_response_parsed_df_3 = pd.json_normalize(data_3['llama_response_parsed'])\n",
        "llama_response_parsed_df_3.head()"
      ],
      "metadata": {
        "id": "NnbnS9dBnj1_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "b7c0973f-8b99-46b3-cb4a-584425803433"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Category                                               Tags\n",
              "0  Technical Issues     [Internet Connectivity, Disconnections, Speed]\n",
              "1   Hardware Issues  [Laptop Won't Start, Urgent Assistance Require...\n",
              "2     Data Recovery  [Data Loss, Essential Documents, Accidental De...\n",
              "3  Technical Issues  [Wi-Fi Connectivity, Signal Strength, Router C...\n",
              "4   Hardware Issues                        [Battery Drain, Smartphone]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6fbc2b66-49d6-40d9-981b-d3c522d8d303\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Won't Start, Urgent Assistance Require...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Data Loss, Essential Documents, Accidental De...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi Connectivity, Signal Strength, Router C...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Drain, Smartphone]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6fbc2b66-49d6-40d9-981b-d3c522d8d303')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-6fbc2b66-49d6-40d9-981b-d3c522d8d303 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-6fbc2b66-49d6-40d9-981b-d3c522d8d303');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-3032d6e1-2680-4890-81c1-8a6536315391\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-3032d6e1-2680-4890-81c1-8a6536315391')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-3032d6e1-2680-4890-81c1-8a6536315391 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "llama_response_parsed_df_3",
              "summary": "{\n  \"name\": \"llama_response_parsed_df_3\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_with_parsed_model_output_3 = pd.concat([data_3, llama_response_parsed_df_3], axis=1)\n",
        "data_with_parsed_model_output_3.head()"
      ],
      "metadata": {
        "id": "bvLdYZO6nu3A",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 293
        },
        "outputId": "e47f0331-68dc-4b40-ad90-17330ad00646"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \\\n",
              "0   Sure! Here is my analysis of your support tic...   \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...   \n",
              "2   {\\n      \"Category\": \"Data Recovery\",\\n      ...   \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...   \n",
              "4   Sure! Here's my analysis of your support tick...   \n",
              "\n",
              "                               llama_response_parsed          Category  \\\n",
              "0  {'Category': 'Technical Issues', 'Tags': ['Int...  Technical Issues   \n",
              "1  {'Category': 'Hardware Issues', 'Tags': ['Lapt...   Hardware Issues   \n",
              "2  {'Category': 'Data Recovery', 'Tags': ['Data L...     Data Recovery   \n",
              "3  {'Category': 'Technical Issues', 'Tags': ['Wi-...  Technical Issues   \n",
              "4  {'Category': 'Hardware Issues', 'Tags': ['Batt...   Hardware Issues   \n",
              "\n",
              "                                                Tags  \n",
              "0     [Internet Connectivity, Disconnections, Speed]  \n",
              "1  [Laptop Won't Start, Urgent Assistance Require...  \n",
              "2  [Data Loss, Essential Documents, Accidental De...  \n",
              "3  [Wi-Fi Connectivity, Signal Strength, Router C...  \n",
              "4                        [Battery Drain, Smartphone]  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e41a082b-bccb-474f-85c5-28cca32ab8a0\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "      <th>llama_response_parsed</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Sure! Here is my analysis of your support tic...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Int...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Lapt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Won't Start, Urgent Assistance Require...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n      \"Category\": \"Data Recovery\",\\n      ...</td>\n",
              "      <td>{'Category': 'Data Recovery', 'Tags': ['Data L...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Data Loss, Essential Documents, Accidental De...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Wi-...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi Connectivity, Signal Strength, Router C...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Here's my analysis of your support tick...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Batt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Drain, Smartphone]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e41a082b-bccb-474f-85c5-28cca32ab8a0')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-e41a082b-bccb-474f-85c5-28cca32ab8a0 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-e41a082b-bccb-474f-85c5-28cca32ab8a0');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-efb5d9c6-df4b-4a6f-ba65-ac3dbc88f71a\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-efb5d9c6-df4b-4a6f-ba65-ac3dbc88f71a')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-efb5d9c6-df4b-4a6f-ba65-ac3dbc88f71a button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_with_parsed_model_output_3",
              "summary": "{\n  \"name\": \"data_with_parsed_model_output_3\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" Sure! Here is my analysis of your support ticket text:\\n    \\n   {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connectivity\\\", \\\"Disconnections\\\", \\\"Speed\\\"]\\n   }\",\n          \" Sure, I can help you classify your support tickets! Based on the information you've given me, here is my prediction for your first support ticket:\\n\\n{\\n  \\\"Category\\\": \\\"Data Recovery\\\",\\n  \\\"Tags\\\": [\\\"USB Drive\\\", \\\"File Recovery\\\", \\\"Crucial Files\\\"]\\n}\\n\\nBased on the information in your support ticket, it seems like you have encountered a data loss issue and are seeking help to recover your crucial files from a formatted USB drive. This falls under the category of Data Recovery, and some relevant tags that could be applied to this ticket include \\\"USB Drive\\\" and \\\"File Recovery\\\".\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Drop\\\", \\\"Frequent\\\"]\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response_parsed\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to drop llama_response and llama_response_parsed variables\n",
        "final_data_3 = data_with_parsed_model_output_3.drop(['llama_response','llama_response_parsed'], axis=1)\n",
        "final_data_3.head()"
      ],
      "metadata": {
        "id": "xI0-DlTBn0xN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "f77b321c-434c-4830-a395-17509927f4a2"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "           Category                                               Tags  \n",
              "0  Technical Issues     [Internet Connectivity, Disconnections, Speed]  \n",
              "1   Hardware Issues  [Laptop Won't Start, Urgent Assistance Require...  \n",
              "2     Data Recovery  [Data Loss, Essential Documents, Accidental De...  \n",
              "3  Technical Issues  [Wi-Fi Connectivity, Signal Strength, Router C...  \n",
              "4   Hardware Issues                        [Battery Drain, Smartphone]  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c0a5b5dc-ee8b-42ab-b16f-a40bd60e708f\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Won't Start, Urgent Assistance Require...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Data Loss, Essential Documents, Accidental De...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi Connectivity, Signal Strength, Router C...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Drain, Smartphone]</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c0a5b5dc-ee8b-42ab-b16f-a40bd60e708f')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-c0a5b5dc-ee8b-42ab-b16f-a40bd60e708f button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-c0a5b5dc-ee8b-42ab-b16f-a40bd60e708f');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-c7db22d2-014c-4a18-b7f2-062d597c0ce2\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-c7db22d2-014c-4a18-b7f2-062d597c0ce2')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-c7db22d2-014c-4a18-b7f2-062d597c0ce2 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_3",
              "summary": "{\n  \"name\": \"final_data_3\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Task 4 - Ticket Categorization, Creating Tags, Assigning Priority, and Returning Structured Output**"
      ],
      "metadata": {
        "id": "717uRFJ3n8q7"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# create a copy of the data\n",
        "data_4 = data.copy()"
      ],
      "metadata": {
        "id": "5QpcwowRgvq3"
      },
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the prompt to get the desired output\n",
        "instruction_4 = \"\"\"\n",
        "    You are an AI analyzing Support Tickets. Classify the support tockets of the provided support ticket text into the following categories:\n",
        "    - Technical Issues\n",
        "    - Data Recovery\n",
        "    - Hardware Issues\n",
        "\n",
        "    Tags should reflect specific problem areas, technologies involved, actions required and support ticket category type.\n",
        "\n",
        "    Determine the Priority level of each ticket based on the issue's urgency and impact on the user's operations. Use the following Priority:\n",
        "    - High\n",
        "    - Medium\n",
        "    - Low\n",
        "\n",
        "    Provide the output in a JSON format with the following keys:\n",
        "    {\n",
        "      \"Category\": \"your_category_prediction\",\n",
        "      \"Tags\": \"your_tags_prediction\",\n",
        "      \"Priority\": \"your_priority_prediction\"\n",
        "    }\n",
        "\n",
        "    Tags should be list item.\n",
        "    Only return the JSON, do NOT return any other text or information.\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "KhyAoasmgxBO"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# complete the code to create a new column llama_response'\n",
        "# by applying the generate_llama_response function to each ticket in the 'support_ticket_text' column of the DataFrame 'data_4'\n",
        "data_4['llama_response'] = data_4['support_ticket_text'].apply(lambda x: generate_llama_response(instruction_4,x))"
      ],
      "metadata": {
        "id": "zzmGLEAZgLCE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ace78067-864a-40ce-b0a5-65f9661284c6"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      26.17 ms /    44 runs   (    0.59 ms per token,  1681.19 tokens per second)\n",
            "llama_print_timings: prompt eval time =     253.31 ms /   284 tokens (    0.89 ms per token,  1121.14 tokens per second)\n",
            "llama_print_timings:        eval time =     731.69 ms /    43 runs   (   17.02 ms per token,    58.77 tokens per second)\n",
            "llama_print_timings:       total time =    1136.91 ms /   327 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      29.20 ms /    52 runs   (    0.56 ms per token,  1780.88 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.78 ms /   283 tokens (    0.83 ms per token,  1210.52 tokens per second)\n",
            "llama_print_timings:        eval time =     861.12 ms /    51 runs   (   16.88 ms per token,    59.22 tokens per second)\n",
            "llama_print_timings:       total time =    1258.94 ms /   334 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      26.40 ms /    46 runs   (    0.57 ms per token,  1742.62 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.49 ms /   279 tokens (    0.84 ms per token,  1194.91 tokens per second)\n",
            "llama_print_timings:        eval time =     760.02 ms /    45 runs   (   16.89 ms per token,    59.21 tokens per second)\n",
            "llama_print_timings:       total time =    1141.53 ms /   324 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.24 ms /    51 runs   (    0.53 ms per token,  1872.45 tokens per second)\n",
            "llama_print_timings: prompt eval time =     234.12 ms /   287 tokens (    0.82 ms per token,  1225.89 tokens per second)\n",
            "llama_print_timings:        eval time =     846.07 ms /    50 runs   (   16.92 ms per token,    59.10 tokens per second)\n",
            "llama_print_timings:       total time =    1239.02 ms /   337 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      30.51 ms /    54 runs   (    0.56 ms per token,  1770.03 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.86 ms /   262 tokens (    0.89 ms per token,  1120.31 tokens per second)\n",
            "llama_print_timings:        eval time =     896.97 ms /    53 runs   (   16.92 ms per token,    59.09 tokens per second)\n",
            "llama_print_timings:       total time =    1303.80 ms /   315 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      22.81 ms /    39 runs   (    0.58 ms per token,  1709.85 tokens per second)\n",
            "llama_print_timings: prompt eval time =     234.00 ms /   262 tokens (    0.89 ms per token,  1119.66 tokens per second)\n",
            "llama_print_timings:        eval time =     646.07 ms /    38 runs   (   17.00 ms per token,    58.82 tokens per second)\n",
            "llama_print_timings:       total time =    1013.16 ms /   300 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      28.13 ms /    47 runs   (    0.60 ms per token,  1670.81 tokens per second)\n",
            "llama_print_timings: prompt eval time =     232.89 ms /   265 tokens (    0.88 ms per token,  1137.90 tokens per second)\n",
            "llama_print_timings:        eval time =     783.00 ms /    46 runs   (   17.02 ms per token,    58.75 tokens per second)\n",
            "llama_print_timings:       total time =    1178.62 ms /   311 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      28.50 ms /    50 runs   (    0.57 ms per token,  1754.69 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.50 ms /   266 tokens (    0.88 ms per token,  1139.21 tokens per second)\n",
            "llama_print_timings:        eval time =     827.85 ms /    49 runs   (   16.89 ms per token,    59.19 tokens per second)\n",
            "llama_print_timings:       total time =    1217.60 ms /   315 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      25.42 ms /    44 runs   (    0.58 ms per token,  1731.06 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.90 ms /   260 tokens (    0.90 ms per token,  1111.61 tokens per second)\n",
            "llama_print_timings:        eval time =     728.78 ms /    43 runs   (   16.95 ms per token,    59.00 tokens per second)\n",
            "llama_print_timings:       total time =    1107.56 ms /   303 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      26.71 ms /    45 runs   (    0.59 ms per token,  1684.95 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.56 ms /   269 tokens (    0.87 ms per token,  1151.73 tokens per second)\n",
            "llama_print_timings:        eval time =     750.22 ms /    44 runs   (   17.05 ms per token,    58.65 tokens per second)\n",
            "llama_print_timings:       total time =    1137.88 ms /   313 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      95.48 ms /   168 runs   (    0.57 ms per token,  1759.48 tokens per second)\n",
            "llama_print_timings: prompt eval time =     234.11 ms /   262 tokens (    0.89 ms per token,  1119.15 tokens per second)\n",
            "llama_print_timings:        eval time =    2838.76 ms /   167 runs   (   17.00 ms per token,    58.83 tokens per second)\n",
            "llama_print_timings:       total time =    3633.02 ms /   429 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      25.82 ms /    47 runs   (    0.55 ms per token,  1820.01 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.08 ms /   266 tokens (    0.88 ms per token,  1141.25 tokens per second)\n",
            "llama_print_timings:        eval time =     776.73 ms /    46 runs   (   16.89 ms per token,    59.22 tokens per second)\n",
            "llama_print_timings:       total time =    1159.03 ms /   312 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.09 ms /    48 runs   (    0.56 ms per token,  1771.94 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.24 ms /   266 tokens (    0.88 ms per token,  1140.46 tokens per second)\n",
            "llama_print_timings:        eval time =     795.09 ms /    47 runs   (   16.92 ms per token,    59.11 tokens per second)\n",
            "llama_print_timings:       total time =    1181.52 ms /   313 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      25.14 ms /    46 runs   (    0.55 ms per token,  1829.97 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.84 ms /   258 tokens (    0.91 ms per token,  1103.34 tokens per second)\n",
            "llama_print_timings:        eval time =     758.49 ms /    45 runs   (   16.86 ms per token,    59.33 tokens per second)\n",
            "llama_print_timings:       total time =    1134.48 ms /   303 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      26.80 ms /    47 runs   (    0.57 ms per token,  1753.60 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.87 ms /   263 tokens (    0.89 ms per token,  1124.56 tokens per second)\n",
            "llama_print_timings:        eval time =     777.54 ms /    46 runs   (   16.90 ms per token,    59.16 tokens per second)\n",
            "llama_print_timings:       total time =    1161.30 ms /   309 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      23.21 ms /    42 runs   (    0.55 ms per token,  1809.56 tokens per second)\n",
            "llama_print_timings: prompt eval time =     204.13 ms /   254 tokens (    0.80 ms per token,  1244.29 tokens per second)\n",
            "llama_print_timings:        eval time =     693.25 ms /    41 runs   (   16.91 ms per token,    59.14 tokens per second)\n",
            "llama_print_timings:       total time =    1030.57 ms /   295 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.83 ms /    50 runs   (    0.56 ms per token,  1796.36 tokens per second)\n",
            "llama_print_timings: prompt eval time =     233.65 ms /   259 tokens (    0.90 ms per token,  1108.49 tokens per second)\n",
            "llama_print_timings:        eval time =     831.32 ms /    49 runs   (   16.97 ms per token,    58.94 tokens per second)\n",
            "llama_print_timings:       total time =    1228.41 ms /   308 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      25.41 ms /    45 runs   (    0.56 ms per token,  1771.03 tokens per second)\n",
            "llama_print_timings: prompt eval time =     205.33 ms /   254 tokens (    0.81 ms per token,  1237.05 tokens per second)\n",
            "llama_print_timings:        eval time =     746.03 ms /    44 runs   (   16.96 ms per token,    58.98 tokens per second)\n",
            "llama_print_timings:       total time =    1098.80 ms /   298 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      24.19 ms /    44 runs   (    0.55 ms per token,  1818.63 tokens per second)\n",
            "llama_print_timings: prompt eval time =     204.18 ms /   255 tokens (    0.80 ms per token,  1248.92 tokens per second)\n",
            "llama_print_timings:        eval time =     725.53 ms /    43 runs   (   16.87 ms per token,    59.27 tokens per second)\n",
            "llama_print_timings:       total time =    1067.29 ms /   298 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      27.21 ms /    50 runs   (    0.54 ms per token,  1837.90 tokens per second)\n",
            "llama_print_timings: prompt eval time =     274.04 ms /   411 tokens (    0.67 ms per token,  1499.79 tokens per second)\n",
            "llama_print_timings:        eval time =     837.61 ms /    49 runs   (   17.09 ms per token,    58.50 tokens per second)\n",
            "llama_print_timings:       total time =    1268.06 ms /   460 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      28.25 ms /    52 runs   (    0.54 ms per token,  1840.90 tokens per second)\n",
            "llama_print_timings: prompt eval time =     391.92 ms /   524 tokens (    0.75 ms per token,  1337.01 tokens per second)\n",
            "llama_print_timings:        eval time =     882.90 ms /    51 runs   (   17.31 ms per token,    57.76 tokens per second)\n",
            "llama_print_timings:       total time =    1444.33 ms /   575 tokens\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Write the code to check the first five rows of the data to confirm whether the new column has been added\n",
        "data_4.head()"
      ],
      "metadata": {
        "id": "uEVDvz3Gg6g_",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "21b1ca38-46fc-45b2-f4df-fcbb3501f910"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \n",
              "0   {\\n      \"Category\": \"Technical Issues\",\\n   ...  \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...  \n",
              "2   {\\n      \"Category\": \"Data Recovery\",\\n      ...  \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...  \n",
              "4   Sure! Here's my analysis of your support tick...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1879f859-abdc-48d1-8a5c-32cae45a6b39\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n      \"Category\": \"Data Recovery\",\\n      ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Here's my analysis of your support tick...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1879f859-abdc-48d1-8a5c-32cae45a6b39')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-1879f859-abdc-48d1-8a5c-32cae45a6b39 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-1879f859-abdc-48d1-8a5c-32cae45a6b39');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-d6fd616a-260a-4eda-8fff-7a9f5adf8a27\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-d6fd616a-260a-4eda-8fff-7a9f5adf8a27')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-d6fd616a-260a-4eda-8fff-7a9f5adf8a27 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_4",
              "summary": "{\n  \"name\": \"data_4\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connectivity\\\", \\\"Disconnections\\\", \\\"Speed\\\"],\\n      \\\"Priority\\\": \\\"High\\\"\\n     }\",\n          \" {\\n      \\\"Category\\\": \\\"Data Recovery\\\",\\n      \\\"Tags\\\": [\\\"USB Drive\\\", \\\"File Recovery\\\", \\\"Crucial Files\\\"],\\n      \\\"Priority\\\": \\\"High\\\"\\n     }\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Frequently Dropping\\\"],\\n      \\\"Priority\\\": \\\"High\\\"\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to apply the extract_json_data function on the llama_response column to create a new column called llama_response_parsed\n",
        "data_4['llama_response_parsed'] = data_4['llama_response'].apply(extract_json_data)\n",
        "data_4['llama_response_parsed'].head()"
      ],
      "metadata": {
        "id": "AhFwgnQlgmti",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a24698a8-156b-4a03-c616-b58d7974181e"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    {'Category': 'Technical Issues', 'Tags': ['Int...\n",
              "1    {'Category': 'Hardware Issues', 'Tags': ['Lapt...\n",
              "2    {'Category': 'Data Recovery', 'Tags': ['Delete...\n",
              "3    {'Category': 'Technical Issues', 'Tags': ['Wi-...\n",
              "4    {'Category': 'Hardware Issues', 'Tags': ['Batt...\n",
              "Name: llama_response_parsed, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to apply the json_normalize on llama_response_parsed variable\n",
        "llama_response_parsed_df_4 = pd.json_normalize(data_4['llama_response_parsed'])\n",
        "llama_response_parsed_df_4.head()"
      ],
      "metadata": {
        "id": "8bDHR4PqhDZ1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "f810581e-77e5-4d6a-81c3-6f663da2eec3"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Category                                               Tags  \\\n",
              "0  Technical Issues     [Internet Connectivity, Disconnections, Speed]   \n",
              "1   Hardware Issues  [Laptop Restart, Hardware Failure, Urgent Assi...   \n",
              "2     Data Recovery    [Deleted Files, Essential Documents, Data Loss]   \n",
              "3  Technical Issues          [Wi-Fi, Signal Strength, Troubleshooting]   \n",
              "4   Hardware Issues                        [Battery Drain, Smartphone]   \n",
              "\n",
              "  Priority  \n",
              "0     High  \n",
              "1     High  \n",
              "2     High  \n",
              "3   Medium  \n",
              "4   Medium  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-abee56e6-b743-42bf-b5b2-86b986544905\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Restart, Hardware Failure, Urgent Assi...</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Deleted Files, Essential Documents, Data Loss]</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi, Signal Strength, Troubleshooting]</td>\n",
              "      <td>Medium</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Drain, Smartphone]</td>\n",
              "      <td>Medium</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-abee56e6-b743-42bf-b5b2-86b986544905')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-abee56e6-b743-42bf-b5b2-86b986544905 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-abee56e6-b743-42bf-b5b2-86b986544905');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-63a6b924-b931-40c3-b62a-b95620ba44c0\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-63a6b924-b931-40c3-b62a-b95620ba44c0')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-63a6b924-b931-40c3-b62a-b95620ba44c0 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "llama_response_parsed_df_4",
              "summary": "{\n  \"name\": \"llama_response_parsed_df_4\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Medium\",\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to concat data_4 and llama_response_parsed_df_4\n",
        "data_with_parsed_model_output_4 = pd.concat([data_4, llama_response_parsed_df_4], axis=1)\n",
        "data_with_parsed_model_output_4.head()"
      ],
      "metadata": {
        "id": "U7Z-trWNhSRX",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 310
        },
        "outputId": "556a1ce3-8a51-4563-f23f-a861054a3176"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \\\n",
              "0   {\\n      \"Category\": \"Technical Issues\",\\n   ...   \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...   \n",
              "2   {\\n      \"Category\": \"Data Recovery\",\\n      ...   \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...   \n",
              "4   Sure! Here's my analysis of your support tick...   \n",
              "\n",
              "                               llama_response_parsed          Category  \\\n",
              "0  {'Category': 'Technical Issues', 'Tags': ['Int...  Technical Issues   \n",
              "1  {'Category': 'Hardware Issues', 'Tags': ['Lapt...   Hardware Issues   \n",
              "2  {'Category': 'Data Recovery', 'Tags': ['Delete...     Data Recovery   \n",
              "3  {'Category': 'Technical Issues', 'Tags': ['Wi-...  Technical Issues   \n",
              "4  {'Category': 'Hardware Issues', 'Tags': ['Batt...   Hardware Issues   \n",
              "\n",
              "                                                Tags Priority  \n",
              "0     [Internet Connectivity, Disconnections, Speed]     High  \n",
              "1  [Laptop Restart, Hardware Failure, Urgent Assi...     High  \n",
              "2    [Deleted Files, Essential Documents, Data Loss]     High  \n",
              "3          [Wi-Fi, Signal Strength, Troubleshooting]   Medium  \n",
              "4                        [Battery Drain, Smartphone]   Medium  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-006b8f26-963d-40fc-a395-688157357f95\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "      <th>llama_response_parsed</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Int...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Lapt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Restart, Hardware Failure, Urgent Assi...</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n      \"Category\": \"Data Recovery\",\\n      ...</td>\n",
              "      <td>{'Category': 'Data Recovery', 'Tags': ['Delete...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Deleted Files, Essential Documents, Data Loss]</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Wi-...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi, Signal Strength, Troubleshooting]</td>\n",
              "      <td>Medium</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Here's my analysis of your support tick...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Batt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Drain, Smartphone]</td>\n",
              "      <td>Medium</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-006b8f26-963d-40fc-a395-688157357f95')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-006b8f26-963d-40fc-a395-688157357f95 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-006b8f26-963d-40fc-a395-688157357f95');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-283925e3-0b09-4414-b604-1b65385f8ce7\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-283925e3-0b09-4414-b604-1b65385f8ce7')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-283925e3-0b09-4414-b604-1b65385f8ce7 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_with_parsed_model_output_4",
              "summary": "{\n  \"name\": \"data_with_parsed_model_output_4\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connectivity\\\", \\\"Disconnections\\\", \\\"Speed\\\"],\\n      \\\"Priority\\\": \\\"High\\\"\\n     }\",\n          \" {\\n      \\\"Category\\\": \\\"Data Recovery\\\",\\n      \\\"Tags\\\": [\\\"USB Drive\\\", \\\"File Recovery\\\", \\\"Crucial Files\\\"],\\n      \\\"Priority\\\": \\\"High\\\"\\n     }\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Frequently Dropping\\\"],\\n      \\\"Priority\\\": \\\"High\\\"\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response_parsed\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Medium\",\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to drop llama_response and llama_response_parsed variables\n",
        "final_data_4 = data_with_parsed_model_output_4.drop(['llama_response','llama_response_parsed'], axis=1)\n",
        "final_data_4.head()"
      ],
      "metadata": {
        "id": "mUZhyHr_hdEN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "6d8aeabe-6001-4e31-af45-a723b5928f8c"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "           Category                                               Tags  \\\n",
              "0  Technical Issues     [Internet Connectivity, Disconnections, Speed]   \n",
              "1   Hardware Issues  [Laptop Restart, Hardware Failure, Urgent Assi...   \n",
              "2     Data Recovery    [Deleted Files, Essential Documents, Data Loss]   \n",
              "3  Technical Issues          [Wi-Fi, Signal Strength, Troubleshooting]   \n",
              "4   Hardware Issues                        [Battery Drain, Smartphone]   \n",
              "\n",
              "  Priority  \n",
              "0     High  \n",
              "1     High  \n",
              "2     High  \n",
              "3   Medium  \n",
              "4   Medium  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-72ec412c-5dc4-4ba2-8650-ed352d13befd\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Restart, Hardware Failure, Urgent Assi...</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Deleted Files, Essential Documents, Data Loss]</td>\n",
              "      <td>High</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi, Signal Strength, Troubleshooting]</td>\n",
              "      <td>Medium</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Drain, Smartphone]</td>\n",
              "      <td>Medium</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-72ec412c-5dc4-4ba2-8650-ed352d13befd')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-72ec412c-5dc4-4ba2-8650-ed352d13befd button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-72ec412c-5dc4-4ba2-8650-ed352d13befd');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-8166e31f-139a-4860-9a3c-7847743ce10b\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-8166e31f-139a-4860-9a3c-7847743ce10b')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-8166e31f-139a-4860-9a3c-7847743ce10b button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_4",
              "summary": "{\n  \"name\": \"final_data_4\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Medium\",\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Task 5 - Ticket Categorization, Creating Tags, Assigning Priority, Assigning ETA, and Returning Structured Output**"
      ],
      "metadata": {
        "id": "nq8Q0e9Np84Q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# create a copy of the data\n",
        "data_5 = data.copy()"
      ],
      "metadata": {
        "id": "e_Xkr8kChkfM"
      },
      "execution_count": 88,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the prompt to get the desired output\n",
        "instruction_5 = \"\"\"\n",
        "    You are an AI analyzing Support Tickets. Classify the support tockets of the provided support ticket text into the following categories:\n",
        "    - Technical Issues\n",
        "    - Data Recovery\n",
        "    - Hardware Issues\n",
        "\n",
        "    Tags should reflect specific problem areas, technologies involved, actions required.\"\n",
        "\n",
        "    Determine the Priority level of each ticket based on the issue's urgency and impact on the user's operations. Use the following Priority:\n",
        "    - High\n",
        "    - Medium\n",
        "    - Low\n",
        "\n",
        "    Assigning ETA based on priority:\n",
        "    - Immediate : if priroity is High\n",
        "    - 24 hours : if priority is Medium\n",
        "    - 2-3 business days : if priority is Low\n",
        "\n",
        "    Provide the output in a JSON format with the following keys:\n",
        "    {\n",
        "      \"Category\": \"your_category_prediction\",\n",
        "      \"Tags\": \"your_tags_prediction\",\n",
        "      \"Priority\": \"your_priority_prediction\",\n",
        "      \"ETA\": \"your_eta_prediction\"\n",
        "    }\n",
        "\n",
        "    Tags should be list item.\n",
        "    Return only return the JSON, no other text or information without JSON.\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "1L-8KbTshkb0"
      },
      "execution_count": 87,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# complete the code to create a new column llama_response'\n",
        "# by applying the generate_llama_response function to each ticket in the 'support_ticket_text' column of the DataFrame 'data_5'\n",
        "data_5['llama_response'] = data_5['support_ticket_text'].apply(lambda x: generate_llama_response(instruction_5,x))"
      ],
      "metadata": {
        "id": "3G0vPnr9hkZ6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "710fa5a2-bbed-4abc-fffa-9ee1f458ab95"
      },
      "execution_count": 89,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      91.56 ms /   154 runs   (    0.59 ms per token,  1681.90 tokens per second)\n",
            "llama_print_timings: prompt eval time =     258.84 ms /   344 tokens (    0.75 ms per token,  1329.03 tokens per second)\n",
            "llama_print_timings:        eval time =    2620.83 ms /   153 runs   (   17.13 ms per token,    58.38 tokens per second)\n",
            "llama_print_timings:       total time =    3412.57 ms /   497 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      32.99 ms /    57 runs   (    0.58 ms per token,  1727.69 tokens per second)\n",
            "llama_print_timings: prompt eval time =     242.04 ms /   343 tokens (    0.71 ms per token,  1417.12 tokens per second)\n",
            "llama_print_timings:        eval time =     953.64 ms /    56 runs   (   17.03 ms per token,    58.72 tokens per second)\n",
            "llama_print_timings:       total time =    1383.56 ms /   399 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      33.20 ms /    58 runs   (    0.57 ms per token,  1746.99 tokens per second)\n",
            "llama_print_timings: prompt eval time =     241.86 ms /   339 tokens (    0.71 ms per token,  1401.64 tokens per second)\n",
            "llama_print_timings:        eval time =     969.77 ms /    57 runs   (   17.01 ms per token,    58.78 tokens per second)\n",
            "llama_print_timings:       total time =    1399.32 ms /   396 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      33.76 ms /    60 runs   (    0.56 ms per token,  1777.25 tokens per second)\n",
            "llama_print_timings: prompt eval time =     241.59 ms /   347 tokens (    0.70 ms per token,  1436.31 tokens per second)\n",
            "llama_print_timings:        eval time =    1008.99 ms /    59 runs   (   17.10 ms per token,    58.47 tokens per second)\n",
            "llama_print_timings:       total time =    1454.42 ms /   406 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =     112.66 ms /   193 runs   (    0.58 ms per token,  1713.20 tokens per second)\n",
            "llama_print_timings: prompt eval time =     239.46 ms /   322 tokens (    0.74 ms per token,  1344.71 tokens per second)\n",
            "llama_print_timings:        eval time =    3283.46 ms /   192 runs   (   17.10 ms per token,    58.47 tokens per second)\n",
            "llama_print_timings:       total time =    4188.72 ms /   514 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      30.17 ms /    52 runs   (    0.58 ms per token,  1723.74 tokens per second)\n",
            "llama_print_timings: prompt eval time =     239.34 ms /   322 tokens (    0.74 ms per token,  1345.36 tokens per second)\n",
            "llama_print_timings:        eval time =     870.72 ms /    51 runs   (   17.07 ms per token,    58.57 tokens per second)\n",
            "llama_print_timings:       total time =    1285.54 ms /   373 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      36.62 ms /    63 runs   (    0.58 ms per token,  1720.56 tokens per second)\n",
            "llama_print_timings: prompt eval time =     239.96 ms /   325 tokens (    0.74 ms per token,  1354.37 tokens per second)\n",
            "llama_print_timings:        eval time =    1054.91 ms /    62 runs   (   17.01 ms per token,    58.77 tokens per second)\n",
            "llama_print_timings:       total time =    1502.86 ms /   387 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      43.88 ms /    73 runs   (    0.60 ms per token,  1663.82 tokens per second)\n",
            "llama_print_timings: prompt eval time =     240.78 ms /   326 tokens (    0.74 ms per token,  1353.95 tokens per second)\n",
            "llama_print_timings:        eval time =    1229.09 ms /    72 runs   (   17.07 ms per token,    58.58 tokens per second)\n",
            "llama_print_timings:       total time =    1717.21 ms /   398 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      31.78 ms /    54 runs   (    0.59 ms per token,  1699.07 tokens per second)\n",
            "llama_print_timings: prompt eval time =     239.17 ms /   320 tokens (    0.75 ms per token,  1337.95 tokens per second)\n",
            "llama_print_timings:        eval time =     901.08 ms /    53 runs   (   17.00 ms per token,    58.82 tokens per second)\n",
            "llama_print_timings:       total time =    1321.87 ms /   373 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      39.59 ms /    68 runs   (    0.58 ms per token,  1717.56 tokens per second)\n",
            "llama_print_timings: prompt eval time =     240.76 ms /   329 tokens (    0.73 ms per token,  1366.53 tokens per second)\n",
            "llama_print_timings:        eval time =    1140.74 ms /    67 runs   (   17.03 ms per token,    58.73 tokens per second)\n",
            "llama_print_timings:       total time =    1604.91 ms /   396 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      25.27 ms /    45 runs   (    0.56 ms per token,  1780.63 tokens per second)\n",
            "llama_print_timings: prompt eval time =     239.75 ms /   322 tokens (    0.74 ms per token,  1343.09 tokens per second)\n",
            "llama_print_timings:        eval time =     748.11 ms /    44 runs   (   17.00 ms per token,    58.81 tokens per second)\n",
            "llama_print_timings:       total time =    1136.02 ms /   366 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      29.20 ms /    51 runs   (    0.57 ms per token,  1746.81 tokens per second)\n",
            "llama_print_timings: prompt eval time =     240.35 ms /   326 tokens (    0.74 ms per token,  1356.37 tokens per second)\n",
            "llama_print_timings:        eval time =     849.21 ms /    50 runs   (   16.98 ms per token,    58.88 tokens per second)\n",
            "llama_print_timings:       total time =    1256.02 ms /   376 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      32.60 ms /    55 runs   (    0.59 ms per token,  1687.01 tokens per second)\n",
            "llama_print_timings: prompt eval time =     240.19 ms /   326 tokens (    0.74 ms per token,  1357.24 tokens per second)\n",
            "llama_print_timings:        eval time =     920.44 ms /    54 runs   (   17.05 ms per token,    58.67 tokens per second)\n",
            "llama_print_timings:       total time =    1345.52 ms /   380 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =     401.60 ms /   694 runs   (    0.58 ms per token,  1728.10 tokens per second)\n",
            "llama_print_timings: prompt eval time =     238.97 ms /   318 tokens (    0.75 ms per token,  1330.69 tokens per second)\n",
            "llama_print_timings:        eval time =   12128.03 ms /   693 runs   (   17.50 ms per token,    57.14 tokens per second)\n",
            "llama_print_timings:       total time =   15068.52 ms /  1011 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      96.52 ms /   163 runs   (    0.59 ms per token,  1688.75 tokens per second)\n",
            "llama_print_timings: prompt eval time =     240.26 ms /   323 tokens (    0.74 ms per token,  1344.39 tokens per second)\n",
            "llama_print_timings:        eval time =    2768.05 ms /   162 runs   (   17.09 ms per token,    58.52 tokens per second)\n",
            "llama_print_timings:       total time =    3573.18 ms /   485 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      28.90 ms /    52 runs   (    0.56 ms per token,  1799.43 tokens per second)\n",
            "llama_print_timings: prompt eval time =     238.64 ms /   314 tokens (    0.76 ms per token,  1315.77 tokens per second)\n",
            "llama_print_timings:        eval time =     865.23 ms /    51 runs   (   16.97 ms per token,    58.94 tokens per second)\n",
            "llama_print_timings:       total time =    1272.75 ms /   365 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =     127.38 ms /   231 runs   (    0.55 ms per token,  1813.43 tokens per second)\n",
            "llama_print_timings: prompt eval time =     239.43 ms /   319 tokens (    0.75 ms per token,  1332.35 tokens per second)\n",
            "llama_print_timings:        eval time =    3936.62 ms /   230 runs   (   17.12 ms per token,    58.43 tokens per second)\n",
            "llama_print_timings:       total time =    4969.42 ms /   549 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      79.89 ms /   143 runs   (    0.56 ms per token,  1789.98 tokens per second)\n",
            "llama_print_timings: prompt eval time =     238.91 ms /   314 tokens (    0.76 ms per token,  1314.29 tokens per second)\n",
            "llama_print_timings:        eval time =    2413.70 ms /   142 runs   (   17.00 ms per token,    58.83 tokens per second)\n",
            "llama_print_timings:       total time =    3120.36 ms /   456 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      30.67 ms /    54 runs   (    0.57 ms per token,  1760.74 tokens per second)\n",
            "llama_print_timings: prompt eval time =     238.72 ms /   315 tokens (    0.76 ms per token,  1319.54 tokens per second)\n",
            "llama_print_timings:        eval time =     898.80 ms /    53 runs   (   16.96 ms per token,    58.97 tokens per second)\n",
            "llama_print_timings:       total time =    1312.23 ms /   368 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      34.24 ms /    60 runs   (    0.57 ms per token,  1752.49 tokens per second)\n",
            "llama_print_timings: prompt eval time =     281.83 ms /   471 tokens (    0.60 ms per token,  1671.22 tokens per second)\n",
            "llama_print_timings:        eval time =    1015.98 ms /    59 runs   (   17.22 ms per token,    58.07 tokens per second)\n",
            "llama_print_timings:       total time =    1509.50 ms /   530 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      29.45 ms /    53 runs   (    0.56 ms per token,  1799.78 tokens per second)\n",
            "llama_print_timings: prompt eval time =     470.18 ms /   584 tokens (    0.81 ms per token,  1242.08 tokens per second)\n",
            "llama_print_timings:        eval time =     902.63 ms /    52 runs   (   17.36 ms per token,    57.61 tokens per second)\n",
            "llama_print_timings:       total time =    1547.57 ms /   636 tokens\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Write the code to check the first five rows of the data to confirm whether the new column has been added\n",
        "data_5.head()"
      ],
      "metadata": {
        "id": "5sz4DW7QhkXp",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "1bd73f09-007f-4880-f634-34c1ba720141"
      },
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \n",
              "0   Sure! Here is the support ticket text you pro...  \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...  \n",
              "2   {\\n      \"Category\": \"Data Recovery\",\\n      ...  \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...  \n",
              "4   Sure! Here's my analysis of your support tick...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-97c5ec8d-1cbb-4c72-9406-d47585e75de2\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Sure! Here is the support ticket text you pro...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n      \"Category\": \"Data Recovery\",\\n      ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Here's my analysis of your support tick...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-97c5ec8d-1cbb-4c72-9406-d47585e75de2')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-97c5ec8d-1cbb-4c72-9406-d47585e75de2 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-97c5ec8d-1cbb-4c72-9406-d47585e75de2');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-73b04baf-8ab4-4378-98f5-085cb36c0dda\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-73b04baf-8ab4-4378-98f5-085cb36c0dda')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-73b04baf-8ab4-4378-98f5-085cb36c0dda button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_5",
              "summary": "{\n  \"name\": \"data_5\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" Sure! Here is the support ticket text you provided, classified into categories and with a predicted priority and ETA:\\n\\n{\\n\\\"Category\\\": \\\"Technical Issues\\\",\\n\\\"Tags\\\": [\\\"Internet Connectivity\\\", \\\"Disconnections\\\", \\\"Speed\\\"],\\n\\\"Priority\\\": \\\"High\\\",\\n\\\"ETA\\\": \\\"Immediate\\\"\\n}\\n\\nBased on the text of the support ticket, I have classified it as a Technical Issues category with tags related to Internet Connectivity and Disconnections. The predicted priority is High, indicating that the issue is urgent and requires immediate attention. The ETA is Immediate, indicating that the support team should prioritize addressing this issue as soon as possible.\",\n          \" Sure, I can help you classify and prioritize your support tickets! Here's the output in JSON format based on the information you provided:\\n\\n{\\n\\\"Category\\\": \\\"Data Recovery\\\",\\n\\\"Tags\\\": [\\\"USB Drive\\\", \\\"Formatted\\\", \\\"Crucial Files\\\"],\\n\\\"Priority\\\": \\\"High\\\",\\n\\\"ETA\\\": \\\"Immediate\\\"\\n}\\n\\nBased on the information in the support ticket, the category is Data Recovery, and the tags are USB Drive, Formatted, and Crucial Files. The priority is High, and the ETA is Immediate, as the user has lost crucial files and needs immediate assistance.\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Frequently Dropping\\\"],\\n      \\\"Priority\\\": \\\"High\\\",\\n      \\\"ETA\\\": \\\"Immediate\\\"\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 90
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to apply the extract_json_data function on the llama_response column to create a new column called llama_response_parsed\n",
        "data_5['llama_response_parsed'] = data_5['llama_response'].apply(extract_json_data)\n",
        "data_5['llama_response_parsed'].head()"
      ],
      "metadata": {
        "id": "3FAqjwbwhkVj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "80d9f01b-1eea-4c8a-c6cb-1586fa248428"
      },
      "execution_count": 91,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Warning: JSON object not found in response:  Sure, I can help you classify and prioritize your support tickets! Please provide the text of the first ticket, and I'll get started:\n",
            "\n",
            "Please provide the text of the first ticket:\n",
            "Error parsing JSON: Extra data: line 5 column 1 (char 124)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    {'Category': 'Technical Issues', 'Tags': ['Int...\n",
              "1    {'Category': 'Hardware Issues', 'Tags': ['Lapt...\n",
              "2    {'Category': 'Data Recovery', 'Tags': ['Delete...\n",
              "3    {'Category': 'Technical Issues', 'Tags': ['Wi-...\n",
              "4    {'Category': 'Hardware Issues', 'Tags': ['Batt...\n",
              "Name: llama_response_parsed, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 91
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to apply the json_normalize on llama_response_parsed variable\n",
        "llama_response_parsed_df_5 = pd.json_normalize(data_5['llama_response_parsed'])\n",
        "llama_response_parsed_df_5.head()"
      ],
      "metadata": {
        "id": "2SRFIpZChkS8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "77b64534-ce47-48a7-ce10-11bb8f35d677"
      },
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Category                                               Tags  \\\n",
              "0  Technical Issues     [Internet Connectivity, Disconnections, Speed]   \n",
              "1   Hardware Issues  [Laptop startup failure, Urgent Assistance Req...   \n",
              "2     Data Recovery  [Deleted Files, Essential Documents, Urgent Re...   \n",
              "3  Technical Issues          [Wi-Fi, Signal Strength, Troubleshooting]   \n",
              "4   Hardware Issues                         [Battery Life, Smartphone]   \n",
              "\n",
              "  Priority        ETA  \n",
              "0     High  Immediate  \n",
              "1     High  Immediate  \n",
              "2     High  Immediate  \n",
              "3     High  Immediate  \n",
              "4   Medium   24 hours  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0dce1f3e-1572-44b3-b358-b61ff6af931b\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "      <th>ETA</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop startup failure, Urgent Assistance Req...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Deleted Files, Essential Documents, Urgent Re...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi, Signal Strength, Troubleshooting]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Life, Smartphone]</td>\n",
              "      <td>Medium</td>\n",
              "      <td>24 hours</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0dce1f3e-1572-44b3-b358-b61ff6af931b')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-0dce1f3e-1572-44b3-b358-b61ff6af931b button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-0dce1f3e-1572-44b3-b358-b61ff6af931b');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-0e136e85-d9c1-4e56-86e7-3c3bf348d8f3\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0e136e85-d9c1-4e56-86e7-3c3bf348d8f3')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-0e136e85-d9c1-4e56-86e7-3c3bf348d8f3 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "llama_response_parsed_df_5",
              "summary": "{\n  \"name\": \"llama_response_parsed_df_5\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Medium\",\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ETA\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"24 hours\",\n          \"Immediate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 92
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to concat data_5 and llama_response_parsed_df_5\n",
        "data_with_parsed_model_output_5 = pd.concat([data_5, llama_response_parsed_df_5], axis=1)\n",
        "data_with_parsed_model_output_5.head()"
      ],
      "metadata": {
        "id": "GL5Pg7dLp_zW",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 379
        },
        "outputId": "8f4733b1-6b75-492d-da36-5c784f908122"
      },
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \\\n",
              "0   Sure! Here is the support ticket text you pro...   \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...   \n",
              "2   {\\n      \"Category\": \"Data Recovery\",\\n      ...   \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...   \n",
              "4   Sure! Here's my analysis of your support tick...   \n",
              "\n",
              "                               llama_response_parsed          Category  \\\n",
              "0  {'Category': 'Technical Issues', 'Tags': ['Int...  Technical Issues   \n",
              "1  {'Category': 'Hardware Issues', 'Tags': ['Lapt...   Hardware Issues   \n",
              "2  {'Category': 'Data Recovery', 'Tags': ['Delete...     Data Recovery   \n",
              "3  {'Category': 'Technical Issues', 'Tags': ['Wi-...  Technical Issues   \n",
              "4  {'Category': 'Hardware Issues', 'Tags': ['Batt...   Hardware Issues   \n",
              "\n",
              "                                                Tags Priority        ETA  \n",
              "0     [Internet Connectivity, Disconnections, Speed]     High  Immediate  \n",
              "1  [Laptop startup failure, Urgent Assistance Req...     High  Immediate  \n",
              "2  [Deleted Files, Essential Documents, Urgent Re...     High  Immediate  \n",
              "3          [Wi-Fi, Signal Strength, Troubleshooting]     High  Immediate  \n",
              "4                         [Battery Life, Smartphone]   Medium   24 hours  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-dd30b50e-408e-41e4-8e9e-2130b30add82\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "      <th>llama_response_parsed</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "      <th>ETA</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Sure! Here is the support ticket text you pro...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Int...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Lapt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop startup failure, Urgent Assistance Req...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>{\\n      \"Category\": \"Data Recovery\",\\n      ...</td>\n",
              "      <td>{'Category': 'Data Recovery', 'Tags': ['Delete...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Deleted Files, Essential Documents, Urgent Re...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Wi-...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi, Signal Strength, Troubleshooting]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Sure! Here's my analysis of your support tick...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Batt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Life, Smartphone]</td>\n",
              "      <td>Medium</td>\n",
              "      <td>24 hours</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-dd30b50e-408e-41e4-8e9e-2130b30add82')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-dd30b50e-408e-41e4-8e9e-2130b30add82 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-dd30b50e-408e-41e4-8e9e-2130b30add82');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-473cd1d2-4cfb-4c1b-96fc-dd26dd6d3273\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-473cd1d2-4cfb-4c1b-96fc-dd26dd6d3273')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-473cd1d2-4cfb-4c1b-96fc-dd26dd6d3273 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_with_parsed_model_output_5",
              "summary": "{\n  \"name\": \"data_with_parsed_model_output_5\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \" Sure! Here is the support ticket text you provided, classified into categories and with a predicted priority and ETA:\\n\\n{\\n\\\"Category\\\": \\\"Technical Issues\\\",\\n\\\"Tags\\\": [\\\"Internet Connectivity\\\", \\\"Disconnections\\\", \\\"Speed\\\"],\\n\\\"Priority\\\": \\\"High\\\",\\n\\\"ETA\\\": \\\"Immediate\\\"\\n}\\n\\nBased on the text of the support ticket, I have classified it as a Technical Issues category with tags related to Internet Connectivity and Disconnections. The predicted priority is High, indicating that the issue is urgent and requires immediate attention. The ETA is Immediate, indicating that the support team should prioritize addressing this issue as soon as possible.\",\n          \" Sure, I can help you classify and prioritize your support tickets! Here's the output in JSON format based on the information you provided:\\n\\n{\\n\\\"Category\\\": \\\"Data Recovery\\\",\\n\\\"Tags\\\": [\\\"USB Drive\\\", \\\"Formatted\\\", \\\"Crucial Files\\\"],\\n\\\"Priority\\\": \\\"High\\\",\\n\\\"ETA\\\": \\\"Immediate\\\"\\n}\\n\\nBased on the information in the support ticket, the category is Data Recovery, and the tags are USB Drive, Formatted, and Crucial Files. The priority is High, and the ETA is Immediate, as the user has lost crucial files and needs immediate assistance.\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Frequently Dropping\\\"],\\n      \\\"Priority\\\": \\\"High\\\",\\n      \\\"ETA\\\": \\\"Immediate\\\"\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response_parsed\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Medium\",\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ETA\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"24 hours\",\n          \"Immediate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 93
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to drop llama_response and llama_response_parsed variables\n",
        "final_data_5 = data_with_parsed_model_output_5.drop(['llama_response','llama_response_parsed'], axis=1)\n",
        "final_data_5.head()"
      ],
      "metadata": {
        "id": "c55UFpTAhsDc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "57536465-d574-4217-bfc4-403a7c45d9ad"
      },
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "           Category                                               Tags  \\\n",
              "0  Technical Issues     [Internet Connectivity, Disconnections, Speed]   \n",
              "1   Hardware Issues  [Laptop startup failure, Urgent Assistance Req...   \n",
              "2     Data Recovery  [Deleted Files, Essential Documents, Urgent Re...   \n",
              "3  Technical Issues          [Wi-Fi, Signal Strength, Troubleshooting]   \n",
              "4   Hardware Issues                         [Battery Life, Smartphone]   \n",
              "\n",
              "  Priority        ETA  \n",
              "0     High  Immediate  \n",
              "1     High  Immediate  \n",
              "2     High  Immediate  \n",
              "3     High  Immediate  \n",
              "4   Medium   24 hours  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-9eb5a22b-a310-4146-9fc8-d97f37e48a30\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "      <th>ETA</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connectivity, Disconnections, Speed]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop startup failure, Urgent Assistance Req...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[Deleted Files, Essential Documents, Urgent Re...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Wi-Fi, Signal Strength, Troubleshooting]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Battery Life, Smartphone]</td>\n",
              "      <td>Medium</td>\n",
              "      <td>24 hours</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-9eb5a22b-a310-4146-9fc8-d97f37e48a30')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-9eb5a22b-a310-4146-9fc8-d97f37e48a30 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-9eb5a22b-a310-4146-9fc8-d97f37e48a30');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-0206fca2-fde1-415e-ade9-ba0a978ba224\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0206fca2-fde1-415e-ade9-ba0a978ba224')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-0206fca2-fde1-415e-ade9-ba0a978ba224 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_5",
              "summary": "{\n  \"name\": \"final_data_5\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"Medium\",\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ETA\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"24 hours\",\n          \"Immediate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 94
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Task 6 - Ticket Categorization, Creating Tags, Assigning Priority, Assigning ETA, Creating a Draft Response, and Returning Structured Output**"
      ],
      "metadata": {
        "id": "y2qmZMOLqG9o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# create a copy of the data\n",
        "data_6 = data.copy()"
      ],
      "metadata": {
        "id": "fsjTgNNoiQtw"
      },
      "execution_count": 137,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the prompt to get the desired output\n",
        "instruction_6 = \"\"\"\n",
        "    You are an AI analyzing Support Tickets. Classify the support tockets of the provided support ticket text into the following categories:\n",
        "    - Technical Issues\n",
        "    - Data Recovery\n",
        "    - Hardware Issues\n",
        "\n",
        "    Tags should reflect specific problem areas, technologies involved or actions required.\n",
        "\n",
        "    Determine the Priority level of each ticket based on the issue's urgency and impact on the user's operations. Use the following Priority:\n",
        "    - High : if support ticket text are like immediate, urgent, ASAP, critical.\n",
        "    - Medium : if support ticket for install or upgrade hardware software.\n",
        "    - Low : other then everything will be low\n",
        "\n",
        "    Assigning ETA based on priority:\n",
        "    - Immediate : if priroity is High\n",
        "    - 24 hours : if priority is Medium\n",
        "    - 2-3 business days : if priority is Low\n",
        "\n",
        "    Finally, draft a response for the customer based on the review. Start out with a thank you note and then add on to it as per the following:\n",
        "    - If the priority is high, We understand the urgency of your issue and are prioritizing its resolution.\n",
        "    - If the priority is medium, We have received your support ticket and understand the importance of resolving your issue promptly.\n",
        "    - If the priority is Low, Your request has been received and is classified as low priority at this moment.\n",
        "\n",
        "    Return the output in the specified JSON format, Ensure that all values in the JSON are formatted as string:\n",
        "\n",
        "    {\n",
        "      \"Category\": \"your_category_prediction\",\n",
        "      \"Tags\": \"your_tags_prediction\",\n",
        "      \"Priority\": \"your_priority_prediction\",\n",
        "      \"ETA\": \"your_eta_prediction\",\n",
        "      \"Response\": \"your_response_to_the_support_ticket\"\n",
        "    }\n",
        "\n",
        "    Be polite and empathetic in the response to the customer review.\n",
        "    Tags should be list item.\n",
        "    Only return the JSON, do NOT return any other text or information.\n",
        "\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "Ky0C2blfiQqZ"
      },
      "execution_count": 138,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# complete the code to create a new column llama_response'\n",
        "# by applying the generate_llama_response function to each ticket in the 'support_ticket_text' column of the DataFrame 'data_6'\n",
        "data_6['llama_response'] = data_6['support_ticket_text'].apply(lambda x: generate_llama_response(instruction_6,x))"
      ],
      "metadata": {
        "id": "vL2ENCXYiQoH",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "85d4b3a7-21b5-42c3-b238-afa430693046"
      },
      "execution_count": 139,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      83.15 ms /   134 runs   (    0.62 ms per token,  1611.51 tokens per second)\n",
            "llama_print_timings: prompt eval time =     436.89 ms /   537 tokens (    0.81 ms per token,  1229.14 tokens per second)\n",
            "llama_print_timings:        eval time =    2345.03 ms /   133 runs   (   17.63 ms per token,    56.72 tokens per second)\n",
            "llama_print_timings:       total time =    3329.39 ms /   670 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      91.05 ms /   146 runs   (    0.62 ms per token,  1603.60 tokens per second)\n",
            "llama_print_timings: prompt eval time =     405.34 ms /   536 tokens (    0.76 ms per token,  1322.35 tokens per second)\n",
            "llama_print_timings:        eval time =    2561.03 ms /   145 runs   (   17.66 ms per token,    56.62 tokens per second)\n",
            "llama_print_timings:       total time =    3569.78 ms /   681 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      78.75 ms /   130 runs   (    0.61 ms per token,  1650.90 tokens per second)\n",
            "llama_print_timings: prompt eval time =     391.49 ms /   532 tokens (    0.74 ms per token,  1358.91 tokens per second)\n",
            "llama_print_timings:        eval time =    2266.98 ms /   129 runs   (   17.57 ms per token,    56.90 tokens per second)\n",
            "llama_print_timings:       total time =    3173.32 ms /   661 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      59.89 ms /    98 runs   (    0.61 ms per token,  1636.28 tokens per second)\n",
            "llama_print_timings: prompt eval time =     422.23 ms /   540 tokens (    0.78 ms per token,  1278.92 tokens per second)\n",
            "llama_print_timings:        eval time =    1714.74 ms /    97 runs   (   17.68 ms per token,    56.57 tokens per second)\n",
            "llama_print_timings:       total time =    2555.12 ms /   637 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      61.57 ms /   102 runs   (    0.60 ms per token,  1656.73 tokens per second)\n",
            "llama_print_timings: prompt eval time =     323.48 ms /   515 tokens (    0.63 ms per token,  1592.04 tokens per second)\n",
            "llama_print_timings:        eval time =    1772.13 ms /   101 runs   (   17.55 ms per token,    56.99 tokens per second)\n",
            "llama_print_timings:       total time =    2497.91 ms /   616 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      64.04 ms /   107 runs   (    0.60 ms per token,  1670.78 tokens per second)\n",
            "llama_print_timings: prompt eval time =     323.14 ms /   515 tokens (    0.63 ms per token,  1593.76 tokens per second)\n",
            "llama_print_timings:        eval time =    1862.79 ms /   106 runs   (   17.57 ms per token,    56.90 tokens per second)\n",
            "llama_print_timings:       total time =    2614.22 ms /   621 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      68.05 ms /   113 runs   (    0.60 ms per token,  1660.47 tokens per second)\n",
            "llama_print_timings: prompt eval time =     333.22 ms /   518 tokens (    0.64 ms per token,  1554.51 tokens per second)\n",
            "llama_print_timings:        eval time =    1964.08 ms /   112 runs   (   17.54 ms per token,    57.02 tokens per second)\n",
            "llama_print_timings:       total time =    2737.00 ms /   630 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      71.32 ms /   116 runs   (    0.61 ms per token,  1626.43 tokens per second)\n",
            "llama_print_timings: prompt eval time =     337.86 ms /   519 tokens (    0.65 ms per token,  1536.14 tokens per second)\n",
            "llama_print_timings:        eval time =    2019.39 ms /   115 runs   (   17.56 ms per token,    56.95 tokens per second)\n",
            "llama_print_timings:       total time =    2815.37 ms /   634 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      65.41 ms /   107 runs   (    0.61 ms per token,  1635.74 tokens per second)\n",
            "llama_print_timings: prompt eval time =     298.28 ms /   512 tokens (    0.58 ms per token,  1716.48 tokens per second)\n",
            "llama_print_timings:        eval time =    1879.73 ms /   107 runs   (   17.57 ms per token,    56.92 tokens per second)\n",
            "llama_print_timings:       total time =    2608.92 ms /   619 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      71.95 ms /   116 runs   (    0.62 ms per token,  1612.32 tokens per second)\n",
            "llama_print_timings: prompt eval time =     356.31 ms /   522 tokens (    0.68 ms per token,  1465.00 tokens per second)\n",
            "llama_print_timings:        eval time =    2025.39 ms /   115 runs   (   17.61 ms per token,    56.78 tokens per second)\n",
            "llama_print_timings:       total time =    2857.87 ms /   637 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      82.61 ms /   137 runs   (    0.60 ms per token,  1658.44 tokens per second)\n",
            "llama_print_timings: prompt eval time =     323.45 ms /   515 tokens (    0.63 ms per token,  1592.23 tokens per second)\n",
            "llama_print_timings:        eval time =    2392.98 ms /   136 runs   (   17.60 ms per token,    56.83 tokens per second)\n",
            "llama_print_timings:       total time =    3266.50 ms /   651 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      71.07 ms /   118 runs   (    0.60 ms per token,  1660.38 tokens per second)\n",
            "llama_print_timings: prompt eval time =     337.32 ms /   519 tokens (    0.65 ms per token,  1538.58 tokens per second)\n",
            "llama_print_timings:        eval time =    2054.33 ms /   117 runs   (   17.56 ms per token,    56.95 tokens per second)\n",
            "llama_print_timings:       total time =    2861.67 ms /   636 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      76.59 ms /   125 runs   (    0.61 ms per token,  1631.98 tokens per second)\n",
            "llama_print_timings: prompt eval time =     337.76 ms /   519 tokens (    0.65 ms per token,  1536.59 tokens per second)\n",
            "llama_print_timings:        eval time =    2180.81 ms /   124 runs   (   17.59 ms per token,    56.86 tokens per second)\n",
            "llama_print_timings:       total time =    3025.98 ms /   643 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      86.93 ms /   143 runs   (    0.61 ms per token,  1645.08 tokens per second)\n",
            "llama_print_timings: prompt eval time =     290.82 ms /   511 tokens (    0.57 ms per token,  1757.08 tokens per second)\n",
            "llama_print_timings:        eval time =    2500.53 ms /   142 runs   (   17.61 ms per token,    56.79 tokens per second)\n",
            "llama_print_timings:       total time =    3370.65 ms /   653 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      74.36 ms /   122 runs   (    0.61 ms per token,  1640.73 tokens per second)\n",
            "llama_print_timings: prompt eval time =     326.76 ms /   516 tokens (    0.63 ms per token,  1579.16 tokens per second)\n",
            "llama_print_timings:        eval time =    2122.62 ms /   121 runs   (   17.54 ms per token,    57.01 tokens per second)\n",
            "llama_print_timings:       total time =    2927.61 ms /   637 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      65.34 ms /   108 runs   (    0.61 ms per token,  1652.79 tokens per second)\n",
            "llama_print_timings: prompt eval time =     288.78 ms /   507 tokens (    0.57 ms per token,  1755.66 tokens per second)\n",
            "llama_print_timings:        eval time =    1877.10 ms /   107 runs   (   17.54 ms per token,    57.00 tokens per second)\n",
            "llama_print_timings:       total time =    2588.29 ms /   614 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      63.81 ms /   110 runs   (    0.58 ms per token,  1723.84 tokens per second)\n",
            "llama_print_timings: prompt eval time =     298.94 ms /   512 tokens (    0.58 ms per token,  1712.70 tokens per second)\n",
            "llama_print_timings:        eval time =    1912.88 ms /   109 runs   (   17.55 ms per token,    56.98 tokens per second)\n",
            "llama_print_timings:       total time =    2640.50 ms /   621 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      78.80 ms /   128 runs   (    0.62 ms per token,  1624.32 tokens per second)\n",
            "llama_print_timings: prompt eval time =     288.52 ms /   507 tokens (    0.57 ms per token,  1757.23 tokens per second)\n",
            "llama_print_timings:        eval time =    2237.22 ms /   127 runs   (   17.62 ms per token,    56.77 tokens per second)\n",
            "llama_print_timings:       total time =    3052.61 ms /   634 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      64.81 ms /   108 runs   (    0.60 ms per token,  1666.33 tokens per second)\n",
            "llama_print_timings: prompt eval time =     288.85 ms /   508 tokens (    0.57 ms per token,  1758.72 tokens per second)\n",
            "llama_print_timings:        eval time =    1876.04 ms /   107 runs   (   17.53 ms per token,    57.04 tokens per second)\n",
            "llama_print_timings:       total time =    2582.92 ms /   615 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      66.09 ms /   112 runs   (    0.59 ms per token,  1694.63 tokens per second)\n",
            "llama_print_timings: prompt eval time =     508.77 ms /   664 tokens (    0.77 ms per token,  1305.11 tokens per second)\n",
            "llama_print_timings:        eval time =    1972.05 ms /   111 runs   (   17.77 ms per token,    56.29 tokens per second)\n",
            "llama_print_timings:       total time =    2921.07 ms /   775 tokens\n",
            "Llama.generate: prefix-match hit\n",
            "\n",
            "llama_print_timings:        load time =     286.40 ms\n",
            "llama_print_timings:      sample time =      58.00 ms /   100 runs   (    0.58 ms per token,  1724.05 tokens per second)\n",
            "llama_print_timings: prompt eval time =     546.61 ms /   777 tokens (    0.70 ms per token,  1421.50 tokens per second)\n",
            "llama_print_timings:        eval time =    1779.68 ms /    99 runs   (   17.98 ms per token,    55.63 tokens per second)\n",
            "llama_print_timings:       total time =    2718.94 ms /   876 tokens\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Write the code to check the first five rows of the data to confirm whether the new column has been added\n",
        "data_6.head()"
      ],
      "metadata": {
        "id": "sViTcXLHiQl-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "f3e53b33-3d35-4a3c-bbe0-525ff3a7bb73"
      },
      "execution_count": 140,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \n",
              "0   {\\n      \"Category\": \"Technical Issues\",\\n   ...  \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...  \n",
              "2   Sure! Here is the output in JSON format based...  \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...  \n",
              "4   {\\n      \"Category\": \"Hardware Issues\",\\n    ...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ae235d9b-7f66-4d8f-b9fc-8ccb6623ca30\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Sure! Here is the output in JSON format based...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ae235d9b-7f66-4d8f-b9fc-8ccb6623ca30')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-ae235d9b-7f66-4d8f-b9fc-8ccb6623ca30 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-ae235d9b-7f66-4d8f-b9fc-8ccb6623ca30');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-70360f6d-9cc3-44c0-97c4-fb11618d0691\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-70360f6d-9cc3-44c0-97c4-fb11618d0691')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-70360f6d-9cc3-44c0-97c4-fb11618d0691 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_6",
              "summary": "{\n  \"name\": \"data_6\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Disconnections\\\", \\\"Frequent Disruptions\\\"],\\n      \\\"Priority\\\": \\\"High\\\",\\n      \\\"ETA\\\": \\\"Immediate\\\",\\n      \\\"Response\\\": \\\"Thank you for reaching out to us with your urgent technical issue. We understand the importance of resolving your internet connection problem as soon as possible and have prioritized its resolution. Our team is working on addressing the issue and restoring your stable internet connection. We will keep you updated on any progress and appreciate your patience during this time.\\\"\\n     }\",\n          \" Sure, I'd be happy to help! Here's the output in JSON format based on the information you've provided:\\n\\n{\\n\\\"Category\\\": \\\"Data Recovery\\\",\\n\\\"Tags\\\": [\\\"USB Drive\\\", \\\"Crucial Files\\\", \\\"Data Recovery\\\"],\\n\\\"Priority\\\": \\\"High\\\",\\n\\\"ETA\\\": \\\"Immediate\\\",\\n\\\"Response\\\": \\\"Thank you for reaching out to us regarding your urgent need for data recovery from your USB drive. We understand the importance of recovering your crucial files as soon as possible and have prioritized your request accordingly.\\\"\\n}\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Frequently Dropping\\\"],\\n      \\\"Priority\\\": \\\"High\\\",\\n      \\\"ETA\\\": \\\"Immediate\\\",\\n      \\\"Response\\\": \\\"Thank you for reaching out to us with your urgent issue. We understand the importance of resolving this matter as soon as possible and have prioritized its resolution. Our team is working on it and we will keep you updated on any progress.\\\"\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 140
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to apply the extract_json_data function on the llama_response column to create a new column called llama_response_parsed\n",
        "data_6['llama_response_parsed'] = data_6['llama_response'].apply(extract_json_data)\n",
        "data_6['llama_response_parsed'].head()"
      ],
      "metadata": {
        "id": "KD4jxJ0oiQj2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "539859ff-d520-40e2-8035-d768cb0bbe8b"
      },
      "execution_count": 141,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    {'Category': 'Technical Issues', 'Tags': ['Int...\n",
              "1    {'Category': 'Hardware Issues', 'Tags': ['Lapt...\n",
              "2    {'Category': 'Data Recovery', 'Tags': ['delete...\n",
              "3    {'Category': 'Technical Issues', 'Tags': ['Wea...\n",
              "4    {'Category': 'Hardware Issues', 'Tags': ['Smar...\n",
              "Name: llama_response_parsed, dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 141
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to apply the normalize on llama_response_parsed variable\n",
        "llama_response_parsed_df_6= pd.json_normalize(data_6['llama_response_parsed'])\n",
        "llama_response_parsed_df_6.head()"
      ],
      "metadata": {
        "id": "RmVEjs4piQha",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "06687eb5-512d-4752-b627-e19280a03b74"
      },
      "execution_count": 142,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "           Category                                               Tags  \\\n",
              "0  Technical Issues  [Internet Connection, Disconnections, Frequent...   \n",
              "1   Hardware Issues   [Laptop Won't Start, Urgent Assistance Required]   \n",
              "2     Data Recovery  [deleted files, essential documents, substanti...   \n",
              "3  Technical Issues              [Weak Wi-Fi Signal, Persistent Issue]   \n",
              "4   Hardware Issues      [Smartphone Battery Drain,  Rapidly Draining]   \n",
              "\n",
              "  Priority        ETA                                           Response  \n",
              "0     High  Immediate  Thank you for reaching out to us with your urg...  \n",
              "1     High  Immediate  Thank you for reaching out to us with your urg...  \n",
              "2     High  Immediate  Thank you for reaching out to us with your urg...  \n",
              "3     High  Immediate  Thank you for reaching out to us regarding you...  \n",
              "4     High  Immediate  Thank you for reaching out to us regarding you...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-52e7d730-a01c-42cc-8db2-558c66b4bd27\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "      <th>ETA</th>\n",
              "      <th>Response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connection, Disconnections, Frequent...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Won't Start, Urgent Assistance Required]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[deleted files, essential documents, substanti...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Weak Wi-Fi Signal, Persistent Issue]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us regarding you...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Smartphone Battery Drain,  Rapidly Draining]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us regarding you...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-52e7d730-a01c-42cc-8db2-558c66b4bd27')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-52e7d730-a01c-42cc-8db2-558c66b4bd27 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-52e7d730-a01c-42cc-8db2-558c66b4bd27');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-c821af58-c2f7-4e78-9aab-85364c542701\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-c821af58-c2f7-4e78-9aab-85364c542701')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-c821af58-c2f7-4e78-9aab-85364c542701 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "llama_response_parsed_df_6",
              "summary": "{\n  \"name\": \"llama_response_parsed_df_6\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ETA\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Immediate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \"Thank you for reaching out to us with your urgent technical issue. We understand the importance of resolving your internet connection problem as soon as possible and have prioritized its resolution. Our team is working on addressing the issue and restoring your stable internet connection. We will keep you updated on any progress and appreciate your patience during this time.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 142
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to concat data_6 and llama_response_parsed_df_6\n",
        "data_with_parsed_model_output_6 = pd.concat([data_6, llama_response_parsed_df_6], axis=1)\n",
        "data_with_parsed_model_output_6.head()"
      ],
      "metadata": {
        "id": "N9dJ5Y-viQfK",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 640
        },
        "outputId": "c3cdea1d-61d0-4b2d-d9e2-0ff4f3e045f7"
      },
      "execution_count": 143,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "                                      llama_response  \\\n",
              "0   {\\n      \"Category\": \"Technical Issues\",\\n   ...   \n",
              "1   {\\n      \"Category\": \"Hardware Issues\",\\n    ...   \n",
              "2   Sure! Here is the output in JSON format based...   \n",
              "3   {\\n      \"Category\": \"Technical Issues\",\\n   ...   \n",
              "4   {\\n      \"Category\": \"Hardware Issues\",\\n    ...   \n",
              "\n",
              "                               llama_response_parsed          Category  \\\n",
              "0  {'Category': 'Technical Issues', 'Tags': ['Int...  Technical Issues   \n",
              "1  {'Category': 'Hardware Issues', 'Tags': ['Lapt...   Hardware Issues   \n",
              "2  {'Category': 'Data Recovery', 'Tags': ['delete...     Data Recovery   \n",
              "3  {'Category': 'Technical Issues', 'Tags': ['Wea...  Technical Issues   \n",
              "4  {'Category': 'Hardware Issues', 'Tags': ['Smar...   Hardware Issues   \n",
              "\n",
              "                                                Tags Priority        ETA  \\\n",
              "0  [Internet Connection, Disconnections, Frequent...     High  Immediate   \n",
              "1   [Laptop Won't Start, Urgent Assistance Required]     High  Immediate   \n",
              "2  [deleted files, essential documents, substanti...     High  Immediate   \n",
              "3              [Weak Wi-Fi Signal, Persistent Issue]     High  Immediate   \n",
              "4      [Smartphone Battery Drain,  Rapidly Draining]     High  Immediate   \n",
              "\n",
              "                                            Response  \n",
              "0  Thank you for reaching out to us with your urg...  \n",
              "1  Thank you for reaching out to us with your urg...  \n",
              "2  Thank you for reaching out to us with your urg...  \n",
              "3  Thank you for reaching out to us regarding you...  \n",
              "4  Thank you for reaching out to us regarding you...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4acaf798-38eb-4f7d-85c1-7792c0c6ef61\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>llama_response</th>\n",
              "      <th>llama_response_parsed</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "      <th>ETA</th>\n",
              "      <th>Response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Int...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connection, Disconnections, Frequent...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Lapt...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Won't Start, Urgent Assistance Required]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Sure! Here is the output in JSON format based...</td>\n",
              "      <td>{'Category': 'Data Recovery', 'Tags': ['delete...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[deleted files, essential documents, substanti...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>{\\n      \"Category\": \"Technical Issues\",\\n   ...</td>\n",
              "      <td>{'Category': 'Technical Issues', 'Tags': ['Wea...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Weak Wi-Fi Signal, Persistent Issue]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us regarding you...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>{\\n      \"Category\": \"Hardware Issues\",\\n    ...</td>\n",
              "      <td>{'Category': 'Hardware Issues', 'Tags': ['Smar...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Smartphone Battery Drain,  Rapidly Draining]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us regarding you...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4acaf798-38eb-4f7d-85c1-7792c0c6ef61')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-4acaf798-38eb-4f7d-85c1-7792c0c6ef61 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-4acaf798-38eb-4f7d-85c1-7792c0c6ef61');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-ef39b0fe-8339-42af-8638-32c1a94daceb\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-ef39b0fe-8339-42af-8638-32c1a94daceb')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-ef39b0fe-8339-42af-8638-32c1a94daceb button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "data_with_parsed_model_output_6",
              "summary": "{\n  \"name\": \"data_with_parsed_model_output_6\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Disconnections\\\", \\\"Frequent Disruptions\\\"],\\n      \\\"Priority\\\": \\\"High\\\",\\n      \\\"ETA\\\": \\\"Immediate\\\",\\n      \\\"Response\\\": \\\"Thank you for reaching out to us with your urgent technical issue. We understand the importance of resolving your internet connection problem as soon as possible and have prioritized its resolution. Our team is working on addressing the issue and restoring your stable internet connection. We will keep you updated on any progress and appreciate your patience during this time.\\\"\\n     }\",\n          \" Sure, I'd be happy to help! Here's the output in JSON format based on the information you've provided:\\n\\n{\\n\\\"Category\\\": \\\"Data Recovery\\\",\\n\\\"Tags\\\": [\\\"USB Drive\\\", \\\"Crucial Files\\\", \\\"Data Recovery\\\"],\\n\\\"Priority\\\": \\\"High\\\",\\n\\\"ETA\\\": \\\"Immediate\\\",\\n\\\"Response\\\": \\\"Thank you for reaching out to us regarding your urgent need for data recovery from your USB drive. We understand the importance of recovering your crucial files as soon as possible and have prioritized your request accordingly.\\\"\\n}\",\n          \" {\\n      \\\"Category\\\": \\\"Technical Issues\\\",\\n      \\\"Tags\\\": [\\\"Internet Connection\\\", \\\"Frequently Dropping\\\"],\\n      \\\"Priority\\\": \\\"High\\\",\\n      \\\"ETA\\\": \\\"Immediate\\\",\\n      \\\"Response\\\": \\\"Thank you for reaching out to us with your urgent issue. We understand the importance of resolving this matter as soon as possible and have prioritized its resolution. Our team is working on it and we will keep you updated on any progress.\\\"\\n     }\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"llama_response_parsed\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ETA\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Immediate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \"Thank you for reaching out to us with your urgent technical issue. We understand the importance of resolving your internet connection problem as soon as possible and have prioritized its resolution. Our team is working on addressing the issue and restoring your stable internet connection. We will keep you updated on any progress and appreciate your patience during this time.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 143
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## Complete the code to drop llama_response and llama_response_parsed variables\n",
        "final_data_6 = data_with_parsed_model_output_6.drop(['llama_response','llama_response_parsed'], axis=1)\n",
        "final_data_6.head()"
      ],
      "metadata": {
        "id": "KpK2-HvriQc4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 293
        },
        "outputId": "eb8dbd05-307d-41d2-8332-c8a5201c3c6f"
      },
      "execution_count": 144,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "  support_tick_id                                support_ticket_text  \\\n",
              "0      ST2023-006  My internet connection has significantly slowe...   \n",
              "1      ST2023-007  Urgent help required! My laptop refuses to sta...   \n",
              "2      ST2023-008  I've accidentally deleted essential work docum...   \n",
              "3      ST2023-009  Despite being in close proximity to my Wi-Fi r...   \n",
              "4      ST2023-010  My smartphone battery is draining rapidly, eve...   \n",
              "\n",
              "           Category                                               Tags  \\\n",
              "0  Technical Issues  [Internet Connection, Disconnections, Frequent...   \n",
              "1   Hardware Issues   [Laptop Won't Start, Urgent Assistance Required]   \n",
              "2     Data Recovery  [deleted files, essential documents, substanti...   \n",
              "3  Technical Issues              [Weak Wi-Fi Signal, Persistent Issue]   \n",
              "4   Hardware Issues      [Smartphone Battery Drain,  Rapidly Draining]   \n",
              "\n",
              "  Priority        ETA                                           Response  \n",
              "0     High  Immediate  Thank you for reaching out to us with your urg...  \n",
              "1     High  Immediate  Thank you for reaching out to us with your urg...  \n",
              "2     High  Immediate  Thank you for reaching out to us with your urg...  \n",
              "3     High  Immediate  Thank you for reaching out to us regarding you...  \n",
              "4     High  Immediate  Thank you for reaching out to us regarding you...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3f2697cb-816b-49bf-a967-20ca7351a0c2\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>support_tick_id</th>\n",
              "      <th>support_ticket_text</th>\n",
              "      <th>Category</th>\n",
              "      <th>Tags</th>\n",
              "      <th>Priority</th>\n",
              "      <th>ETA</th>\n",
              "      <th>Response</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ST2023-006</td>\n",
              "      <td>My internet connection has significantly slowe...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Internet Connection, Disconnections, Frequent...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ST2023-007</td>\n",
              "      <td>Urgent help required! My laptop refuses to sta...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Laptop Won't Start, Urgent Assistance Required]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ST2023-008</td>\n",
              "      <td>I've accidentally deleted essential work docum...</td>\n",
              "      <td>Data Recovery</td>\n",
              "      <td>[deleted files, essential documents, substanti...</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us with your urg...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ST2023-009</td>\n",
              "      <td>Despite being in close proximity to my Wi-Fi r...</td>\n",
              "      <td>Technical Issues</td>\n",
              "      <td>[Weak Wi-Fi Signal, Persistent Issue]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us regarding you...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ST2023-010</td>\n",
              "      <td>My smartphone battery is draining rapidly, eve...</td>\n",
              "      <td>Hardware Issues</td>\n",
              "      <td>[Smartphone Battery Drain,  Rapidly Draining]</td>\n",
              "      <td>High</td>\n",
              "      <td>Immediate</td>\n",
              "      <td>Thank you for reaching out to us regarding you...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3f2697cb-816b-49bf-a967-20ca7351a0c2')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-3f2697cb-816b-49bf-a967-20ca7351a0c2 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-3f2697cb-816b-49bf-a967-20ca7351a0c2');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-570cc38c-3f86-4151-9486-06ee5e0ecfed\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-570cc38c-3f86-4151-9486-06ee5e0ecfed')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http:https://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-570cc38c-3f86-4151-9486-06ee5e0ecfed button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "variable_name": "final_data_6",
              "summary": "{\n  \"name\": \"final_data_6\",\n  \"rows\": 21,\n  \"fields\": [\n    {\n      \"column\": \"support_tick_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"ST2023-006\",\n          \"ST2023-023\",\n          \"ST2023-021\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"support_ticket_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 21,\n        \"samples\": [\n          \"My internet connection has significantly slowed down over the past two days, making it challenging to work efficiently from home. Frequent disconnections are causing major disruptions. Please assist in resolving this connectivity issue promptly.\",\n          \"I accidentally formatted my USB drive with crucial files. Help needed for data recovery.\",\n          \"My internet connection is frequently dropping, affecting my work. Urgent help needed.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Technical Issues\",\n          \"Hardware Issues\",\n          \"Data Recovery\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Tags\",\n      \"properties\": {\n        \"dtype\": \"object\",\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Priority\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"High\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"ETA\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Immediate\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"Response\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 20,\n        \"samples\": [\n          \"Thank you for reaching out to us with your urgent technical issue. We understand the importance of resolving your internet connection problem as soon as possible and have prioritized its resolution. Our team is working on addressing the issue and restoring your stable internet connection. We will keep you updated on any progress and appreciate your patience during this time.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 144
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Model Output Analysis**"
      ],
      "metadata": {
        "id": "PdivXxvYYCqN"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# creating a copy of the dataframe\n",
        "final_data = final_data_6.copy()"
      ],
      "metadata": {
        "id": "jyvWARc-jjA9"
      },
      "execution_count": 145,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "final_data['Category'].value_counts()   # complete the code to check the distribution of categories"
      ],
      "metadata": {
        "id": "LMjhi2z8ji9l",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "353a7b92-96f3-4231-dcc5-64099674e2d1"
      },
      "execution_count": 146,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Technical Issues    8\n",
              "Hardware Issues     7\n",
              "Data Recovery       6\n",
              "Name: Category, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 146
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data['Priority'].value_counts()   # complete the code to check the distribution of priority"
      ],
      "metadata": {
        "id": "ozKA98E-ji07",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f3c288b5-81b5-495d-f26d-0e0a0dcd5b5b"
      },
      "execution_count": 147,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "High    21\n",
              "Name: Priority, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 147
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data['ETA'].value_counts()   # complete the code to check the distribution of ETA"
      ],
      "metadata": {
        "id": "MMEmUy12kGjg",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "557632f9-4a2d-4617-98fa-56d2ce0add84"
      },
      "execution_count": 148,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Immediate    21\n",
              "Name: ETA, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 148
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data.groupby(['Category', 'Priority']).support_tick_id.count()   # complete the code to check the distribution of priority by categories"
      ],
      "metadata": {
        "id": "wRaN-O8bkGgS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6a95d615-07f2-4af7-8c3f-46a49b114861"
      },
      "execution_count": 149,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Category          Priority\n",
              "Data Recovery     High        6\n",
              "Hardware Issues   High        7\n",
              "Technical Issues  High        8\n",
              "Name: support_tick_id, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 149
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_data.groupby(['Category', 'ETA']).support_tick_id.count()   # complete the code to check the distribution of ETA by categories"
      ],
      "metadata": {
        "id": "7AMkF7QUqUaV",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8bce03ce-3837-4bc3-d6dd-3c31fa729208"
      },
      "execution_count": 150,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Category          ETA      \n",
              "Data Recovery     Immediate    6\n",
              "Hardware Issues   Immediate    7\n",
              "Technical Issues  Immediate    8\n",
              "Name: support_tick_id, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 150
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Observations**\n",
        "\n",
        "- Among the 21 support tickets analyzed, the breakdown is as follows: 8 tickets are related to Technical Issues, 7 to Hardware Issues, and 6 to Data Recovery.\n",
        "- Every ticket has been assigned a High priority level.\n",
        "- The Estimated Time of Arrival (ETA) for addressing each ticket is categorized as Immediate.\n",
        "\n"
      ],
      "metadata": {
        "id": "tBKsy2o1kOHq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## **Actionable Insights and Recommendations**"
      ],
      "metadata": {
        "id": "iVq4hE0FqmEr"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Utilizing a Large Language Model (LLM), we systematically processed technical support tickets through a multi-step approach, focusing on categorization, tagging, priority assignment, ETA determination, and crafting responses:\n",
        "\n",
        "- Our initial step involved classifying each ticket into predefined categories such as Technical Issues, Hardware Issues, or Data Recovery, using the LLM's capabilities.\n",
        "- We further analyzed the content to assign specific tags that reflect problem areas, technologies involved, or actions required, facilitating easier access and organization of the tickets.\n",
        "- The process then involved assessing the urgency and impact of each issue to assign a priority level (High, Medium, Low) to every ticket.\n",
        "- Based on the assigned priority, we determined an Estimated Time of Arrival (ETA) for addressing the ticket, aligning with predefined criteria (Immediate for High priority, 24 hours for Medium, and 2-3 business days for Low priority).\n",
        "- Finally, leveraging all the above analyses, we generated a response tailored to each customer's issue, ready to be shared for communication.\n",
        "\n",
        "For an objective assessment of the model's performance in these tasks, one could manually review a subset of tickets to evaluate the accuracy of category assignments, tag accuracy, priority levels, ETA appropriateness, and the relevance of generated responses. This manual review serves as a benchmark to quantify the model's effectiveness.\n",
        "\n",
        "To further refine the model's output and enhance its performance in handling technical support tickets, consider the following strategies:\n",
        "\n",
        "- Revise the input prompts to ensure they accurately capture the nuances required for categorizing, tagging, and prioritizing tickets.\n",
        "- Adjust the model's operational parameters, such as 'temperature' and 'top_p', to fine-tune its decision-making process and response generation capabilities, aiming for improved accuracy and relevance in its outputs."
      ],
      "metadata": {
        "id": "lwt00vrYqpRo"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "---\n",
        "\n"
      ],
      "metadata": {
        "id": "hrCBAnP-qsqS"
      }
    }
  ]
}