fltlnd.ipynb

{
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.10.12",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "accelerator": "GPU"
  },
  "nbformat_minor": 0,
  "nbformat": 4,
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "## Data import"
      ],
      "metadata": {
        "id": "2Hq7fKmt2pSh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "data = np.load('/content/drive/MyDrive/Colab Notebooks/Duomenų folderis/flatland_train.npz') #Collab\n",
        "#data = np.load('C:\\\\Users\\\\mazas\\\\Documents\\\\_AI_Primer\\\\flatland_train.npz') #PC\n",
        "#data = np.load('/kaggle/input/flatland-train/flatland_train.npz') #Kaggle\n",
        "X = data['X']\n",
        "y = data['y']\n",
        "\n",
        "y[y != 0] -= 2    # Correct labels so that triangle is mapped to class 1\n",
        "X = X / 255"
      ],
      "metadata": {
        "id": "0NkrWOF9proM",
        "outputId": "0eeb126f-8732-4cd0-e79e-332061a17337",
        "execution": {
          "iopub.status.busy": "2023-11-09T11:03:22.775381Z",
          "iopub.execute_input": "2023-11-09T11:03:22.776637Z",
          "iopub.status.idle": "2023-11-09T11:03:23.460950Z",
          "shell.execute_reply.started": "2023-11-09T11:03:22.776582Z",
          "shell.execute_reply": "2023-11-09T11:03:23.459396Z"
        },
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from fltlnd_imports import *"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2023-11-09T11:03:24.865097Z",
          "iopub.execute_input": "2023-11-09T11:03:24.866082Z",
          "iopub.status.idle": "2023-11-09T11:03:25.605836Z",
          "shell.execute_reply.started": "2023-11-09T11:03:24.866035Z",
          "shell.execute_reply": "2023-11-09T11:03:25.604627Z"
        },
        "trusted": true,
        "id": "tevZ3z-pw88k"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Train/Val/Test split"
      ],
      "metadata": {
        "id": "19wtlDKehTa2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_val, X_test, y_train, y_val, y_test = train_val_test_split(X, y, random_state=42)"
      ],
      "metadata": {
        "id": "xDGjpKFghPYW",
        "execution": {
          "iopub.status.busy": "2023-11-09T11:04:01.412485Z",
          "iopub.execute_input": "2023-11-09T11:04:01.413169Z",
          "iopub.status.idle": "2023-11-09T11:04:01.605970Z",
          "shell.execute_reply.started": "2023-11-09T11:04:01.413122Z",
          "shell.execute_reply": "2023-11-09T11:04:01.604330Z"
        },
        "trusted": true
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Augmentation using older keras"
      ],
      "metadata": {
        "id": "7AVdAQjznrhr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Create an ImageDataGenerator with some augmentations\n",
        "datagen = ImageDataGenerator(shear_range=20,\n",
        "                             fill_mode = 'constant',\n",
        "                             cval = 0,\n",
        "                             rotation_range=90,\n",
        "                             zoom_range=(0.7,1.3),\n",
        "                             #brightness_range=(0.999999,0.99999999),\n",
        "                             horizontal_flip = True,\n",
        "                             vertical_flip = True\n",
        "                             #rescale = 1./255\n",
        "                             )"
      ],
      "metadata": {
        "id": "JueDyffPqpvn"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_aug, y_aug = augment(X_train, y_train, datagen)"
      ],
      "metadata": {
        "id": "LHQ4_fSArN5J"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot10(X_aug, y_aug,40)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 525
        },
        "id": "3wBwRUWlreOw",
        "outputId": "915bfc7b-6144-4ed4-cf65-5f7f97a2f02a"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 10 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data label encoding and mixup"
      ],
      "metadata": {
        "id": "1tBcP9VLOOpT"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Convert labels to one-hot encoded vectors\n",
        "y_train = to_categorical(y_train, num_classes=5)\n",
        "y_val = to_categorical(y_val, num_classes=5)\n",
        "y_test = to_categorical(y_test, num_classes=5)\n",
        "y_aug = to_categorical(y_aug, num_classes=5)"
      ],
      "metadata": {
        "id": "Lbn6qKkZONZm",
        "trusted": true
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Data mixup"
      ],
      "metadata": {
        "id": "KDZj-zZSw88p"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X_mix, y_mix = mixup_data(X_aug, y_aug, 0.5, size = 0.4)"
      ],
      "metadata": {
        "trusted": true,
        "id": "3Rdlk8HKw88p"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot10(X_mix, np.round(y_mix,1),10)"
      ],
      "metadata": {
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 525
        },
        "id": "Qjzj8hCDw88q",
        "outputId": "627784d8-1534-4ab9-c0f8-0fa19ac81085"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 10 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#X_temp1, y_temp1 = merge_and_shuffle_for_one_hot(X_train, y_train, X_mix, y_mix)\n",
        "X_train, y_train = merge_and_shuffle_for_one_hot(X_mix, y_mix, X_aug, y_aug)"
      ],
      "metadata": {
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pAmI9oxfw88q",
        "outputId": "8f837ec3-600a-4343-8ccd-3dbb2dbdb0a6"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(8400, 2500)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot10(X_train, np.round(y_train,1),60)"
      ],
      "metadata": {
        "trusted": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 525
        },
        "id": "UALUtyrZw88q",
        "outputId": "440446c8-cf00-4116-ef89-6e3236fea622"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x600 with 10 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model"
      ],
      "metadata": {
        "id": "zsBazI6Vho8l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model = Sequential([\n",
        "\n",
        "    # Convolutional Block 1\n",
        "    Conv2D(10, (3, 3), padding = 'same', activation='relu', kernel_initializer='he_uniform', input_shape=(50, 50, 1)),\n",
        "    BatchNormalization(),\n",
        "    MaxPooling2D(pool_size=(2, 2)),\n",
        "    Dropout(0.2),\n",
        "\n",
        "    # Convolutional Block 2\n",
        "    Conv2D(10*2, (3, 3), padding = 'same', activation='relu', kernel_initializer='he_uniform', input_shape=(50, 50, 1)),\n",
        "    BatchNormalization(),\n",
        "    MaxPooling2D(pool_size=(2, 2)),\n",
        "    Dropout(0.2),\n",
        "\n",
        "    # Convolutional Block 3\n",
        "    Conv2D(10*4, (3, 3), padding = 'same', activation='relu', kernel_initializer='he_uniform', input_shape=(50, 50, 1)),\n",
        "    BatchNormalization(),\n",
        "    MaxPooling2D(pool_size=(2, 2)),\n",
        "    Dropout(0.2),\n",
        "\n",
        "    # Convolutional Block 4\n",
        "    Conv2D(10*8, (3, 3), padding = 'same', activation='relu', kernel_initializer='he_uniform', input_shape=(50, 50, 1)),\n",
        "    BatchNormalization(),\n",
        "    MaxPooling2D(pool_size=(2, 2)),\n",
        "    Dropout(0.2),\n",
        "\n",
        "    # Fully Connected Layers\n",
        "    Flatten(),\n",
        "    Dense(32, activation='relu'),\n",
        "    BatchNormalization(),\n",
        "    Dropout(0.2),\n",
        "    Dense(16, activation='relu'),\n",
        "    Dropout(0.2),\n",
        "    Dense(5, activation='softmax')\n",
        "])\n",
        "\n",
        "# Compile the model\n",
        "model.compile(loss='categorical_crossentropy',\n",
        "              optimizer=Adam(learning_rate=3e-4))\n"
      ],
      "metadata": {
        "id": "1FXrovRkpG9A",
        "trusted": true
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#model.summary()"
      ],
      "metadata": {
        "trusted": true,
        "id": "RQMDcSIQw88t"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_and_eval_pipe(model, X_train, y_train, X_val, y_val, epochs_number = 350)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 507
        },
        "id": "Ywt5uqCkxhAW",
        "outputId": "cca8a084-8f13-4892-f90a-10af75a02723"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train set accuracy: 0.96178573\n",
            "Validation set accuracy: 0.988\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_set_accuracy(model, X_test, y_test)"
      ],
      "metadata": {
        "id": "VnaUxB52qoDL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "030ca763-7c01-4417-8ded-27cd4dbd62b2"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test set accuracy: 0.984\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#model.save('model115_2.h5')"
      ],
      "metadata": {
        "id": "fglbmYFv2Au1"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}