examples/ontology_creator.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "760a4edc-7cac-45a8-af8b-719f32b1a3cb",
   "metadata": {},
   "source": [
    "# Inferring labels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e3c18d4-b287-4831-9a8f-742d0be3c34b",
   "metadata": {},
   "source": [
    "This notebook shows how to construct reasoning chains to infer the target label set. It is helpful when you don't know which ontology might be presented in the dataset of interest. The typical approach is to look at each individual example and add a new category to the ontology when met. But this manual approach can be automated. \n",
    "\n",
    "To give an example, let's load a dataset of news:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1e87ea4a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Wall St. Bears Claw Back Into the Black (Reute...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Carlyle Looks Toward Commercial Aerospace (Reu...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Oil and Economy Cloud Stocks' Outlook (Reuters...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Iraq Halts Oil Exports from Main Southern Pipe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Oil prices soar to all-time record, posing new...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>U.S. Stocks Rebound as Oil Prices Ease  NEW YO...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>Dollar Rises Vs Euro After Asset Data  NEW YOR...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>Bikes Bring Internet to Indian Villagers (AP) ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>Celebrity Chefs Are Everywhere in Vegas By ADA...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>Entertainment World Wary of Microsoft Technolo...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                  text\n",
       "0    Wall St. Bears Claw Back Into the Black (Reute...\n",
       "1    Carlyle Looks Toward Commercial Aerospace (Reu...\n",
       "2    Oil and Economy Cloud Stocks' Outlook (Reuters...\n",
       "3    Iraq Halts Oil Exports from Main Southern Pipe...\n",
       "4    Oil prices soar to all-time record, posing new...\n",
       "..                                                 ...\n",
       "995  U.S. Stocks Rebound as Oil Prices Ease  NEW YO...\n",
       "996  Dollar Rises Vs Euro After Asset Data  NEW YOR...\n",
       "997  Bikes Bring Internet to Indian Villagers (AP) ...\n",
       "998  Celebrity Chefs Are Everywhere in Vegas By ADA...\n",
       "999  Entertainment World Wary of Microsoft Technolo...\n",
       "\n",
       "[1000 rows x 1 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from datasets import load_dataset\n",
    "\n",
    "ds = load_dataset('ag_news', 'default')\n",
    "df = pd.DataFrame(data=ds['train'][:1000]['text'], columns=['text'])\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d474d957-28e9-4f00-9d14-c83d5fa379ff",
   "metadata": {},
   "source": [
    "# Label creation agent\n",
    "\n",
    "The first agent receives raw texts from the news dataset and determines the best label set for the given target. \n",
    "\n",
    "For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f220fc43-66cf-4110-b6b8-2c4bd8138d21",
   "metadata": {},
   "outputs": [],
   "source": [
    "target = 'help analysts identify the primary reason for changes in the stock market.'"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fd2a3f81-4a5e-42fd-8648-8d1290e81812",
   "metadata": {},
   "source": [
    "To create the possible label set that fits this target, agent's reasoning path consists of 2 stages:\n",
    "\n",
    "1. `onto_creator` - create possible candidates for labels;\n",
    "2. `onto_merger` - merge label candidates with similar semantics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0348e01a-7493-446e-96e4-5ea08f41150f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Applying skill: onto_creator\n",
       "</pre>\n"
      ],
      "text/plain": [
       "Applying skill: onto_creator\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing chunks: 100%|██████████████████████| 5/5 [00:20<00:00,  4.12s/it]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">                                         \n",
       " <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\"> categories                            </span> \n",
       " ─────────────────────────────────────── \n",
       "  Economic Factors                       \n",
       "  Market Trends                          \n",
       "  Corporate News and Events              \n",
       "  Commodity Prices                       \n",
       "  Geopolitical Events                    \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Market Trends and Performance         </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Technological Innovations             </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Consumer Behavior                     </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Corporate Strategy                    </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Legal and Regulatory Issues           </span> \n",
       "  Technology and Computing               \n",
       "  Economics and Business                 \n",
       "  Legal and Crimes                       \n",
       "  Politics                               \n",
       "  Science and Environment                \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Macro-economic Factors                </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Geopolitical Events                   </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Natural Disasters                     </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Company News                          </span> \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Investment &amp; Trading Activities       </span> \n",
       "  Corporate Earnings                     \n",
       "  Commodity Prices                       \n",
       "  Mergers and Acquisitions               \n",
       "  Natural Disasters and Weather Events   \n",
       "  Technological Innovations and Updates  \n",
       "                                         \n",
       "</pre>\n"
      ],
      "text/plain": [
       "                                         \n",
       " \u001b[1;35m \u001b[0m\u001b[1;35mcategories                           \u001b[0m\u001b[1;35m \u001b[0m \n",
       " ─────────────────────────────────────── \n",
       "  Economic Factors                       \n",
       "  Market Trends                          \n",
       "  Corporate News and Events              \n",
       "  Commodity Prices                       \n",
       "  Geopolitical Events                    \n",
       " \u001b[2m \u001b[0m\u001b[2mMarket Trends and Performance        \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mTechnological Innovations            \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mConsumer Behavior                    \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mCorporate Strategy                   \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mLegal and Regulatory Issues          \u001b[0m\u001b[2m \u001b[0m \n",
       "  Technology and Computing               \n",
       "  Economics and Business                 \n",
       "  Legal and Crimes                       \n",
       "  Politics                               \n",
       "  Science and Environment                \n",
       " \u001b[2m \u001b[0m\u001b[2mMacro-economic Factors               \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mGeopolitical Events                  \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mNatural Disasters                    \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mCompany News                         \u001b[0m\u001b[2m \u001b[0m \n",
       " \u001b[2m \u001b[0m\u001b[2mInvestment & Trading Activities      \u001b[0m\u001b[2m \u001b[0m \n",
       "  Corporate Earnings                     \n",
       "  Commodity Prices                       \n",
       "  Mergers and Acquisitions               \n",
       "  Natural Disasters and Weather Events   \n",
       "  Technological Innovations and Updates  \n",
       "                                         \n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Applying skill: onto_merger\n",
       "</pre>\n"
      ],
      "text/plain": [
       "Applying skill: onto_merger\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Processing chunks: 1it [00:03,  3.50s/it]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">                               \n",
       " <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\"> labels                      </span> \n",
       " ───────────────────────────── \n",
       "  Economic Factors             \n",
       "  Market Trends                \n",
       "  Corporate News and Events    \n",
       "  Commodity Prices             \n",
       "  Geopolitical Events          \n",
       "  Technological Innovations    \n",
       "  Consumer Behavior            \n",
       "  Corporate Strategy           \n",
       "  Legal and Regulatory Issues  \n",
       "  Natural Disasters            \n",
       "                               \n",
       "</pre>\n"
      ],
      "text/plain": [
       "                               \n",
       " \u001b[1;35m \u001b[0m\u001b[1;35mlabels                     \u001b[0m\u001b[1;35m \u001b[0m \n",
       " ───────────────────────────── \n",
       "  Economic Factors             \n",
       "  Market Trends                \n",
       "  Corporate News and Events    \n",
       "  Commodity Prices             \n",
       "  Geopolitical Events          \n",
       "  Technological Innovations    \n",
       "  Consumer Behavior            \n",
       "  Corporate Strategy           \n",
       "  Legal and Regulatory Issues  \n",
       "  Natural Disasters            \n",
       "                               \n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Economic Factors', 'Market Trends', 'Corporate News and Events', 'Commodity Prices', 'Geopolitical Events', 'Technological Innovations', 'Consumer Behavior', 'Corporate Strategy', 'Legal and Regulatory Issues', 'Natural Disasters']\n"
     ]
    }
   ],
   "source": [
    "from adala.agents import Agent\n",
    "from adala.runtimes import OpenAIChatRuntime\n",
    "from adala.skills import OntologyCreator, OntologyMerger, LinearSkillSet\n",
    "\n",
    "a = Agent(\n",
    "    skills=LinearSkillSet(skills=[\n",
    "        OntologyCreator(target=target),\n",
    "        OntologyMerger(target=target)\n",
    "    ]),\n",
    "    runtimes={\n",
    "        'gpt4': OpenAIChatRuntime(model='gpt-4-1106-preview', verbose=False)\n",
    "    },\n",
    "    default_runtime='gpt4')\n",
    "\n",
    "pred = a.run(df)\n",
    "labels = [l.strip() for l in pred.labels[0].split('\\n')]\n",
    "print(labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f9ba0a69-75cf-430e-bc27-7306c0480ac5",
   "metadata": {},
   "source": [
    "# Classification agent\n",
    "\n",
    "Now, given the label set, we can run classification skills to obtain the prediction for each news"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ef0b0408-4a43-4c23-a65b-05819cefe4c3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">Applying skill: classification\n",
       "</pre>\n"
      ],
      "text/plain": [
       "Applying skill: classification\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████| 100/100 [01:31<00:00,  1.09it/s]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">                         \n",
       " <span style=\"color: #800080; text-decoration-color: #800080; font-weight: bold\"> label                 </span> \n",
       " ─────────────────────── \n",
       "  Corporate Performance  \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Product Announcements </span> \n",
       "  Product Announcements  \n",
       " <span style=\"color: #7f7f7f; text-decoration-color: #7f7f7f\"> Technology Trends     </span> \n",
       "  Regulatory Policy      \n",
       "                         \n",
       "</pre>\n"
      ],
      "text/plain": [
       "                         \n",
       " \u001b[1;35m \u001b[0m\u001b[1;35mlabel                \u001b[0m\u001b[1;35m \u001b[0m \n",
       " ─────────────────────── \n",
       "  Corporate Performance  \n",
       " \u001b[2m \u001b[0m\u001b[2mProduct Announcements\u001b[0m\u001b[2m \u001b[0m \n",
       "  Product Announcements  \n",
       " \u001b[2m \u001b[0m\u001b[2mTechnology Trends    \u001b[0m\u001b[2m \u001b[0m \n",
       "  Regulatory Policy      \n",
       "                         \n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Axes: ylabel='count'>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from adala.skills import ClassificationSkill\n",
    "\n",
    "predictor = Agent(\n",
    "    skills=LinearSkillSet(skills=[ClassificationSkill(labels={'label': labels})]),\n",
    "    runtimes={\n",
    "        'default': OpenAIChatRuntime(model='gpt-4-1106-preview', verbose=False)\n",
    "    })\n",
    "\n",
    "df_test = pd.DataFrame(data=ds['test'][:100]['text'], columns=['text'])\n",
    "\n",
    "pred_test = predictor.run(df_test)\n",
    "pred_test.label.value_counts().plot.pie(figsize=(12, 12))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "adala-pdm",
   "language": "python",
   "name": "adala-pdm"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}