Risk_Category_TabNet_IMBALANCE_76ACC.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyMi+IZUrw4RTzlsBhG2pjCI",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/fjadidi2001/Insurance/blob/main/Risk_Category_TabNet.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E1m4-kqHuEge",
        "outputId": "20353317-4eb0-4e3d-cc9f-b46473fb6d12"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n",
            "(100000, 52)\n",
            "   Duration  Insured.age Insured.sex  Car.age  Marital  Car.use  Credit.score  \\\n",
            "0       366           45        Male       -1  Married  Commute         609.0   \n",
            "1       182           44      Female        3  Married  Commute         575.0   \n",
            "2       184           48      Female        6  Married  Commute         847.0   \n",
            "3       183           71        Male        6  Married  Private         842.0   \n",
            "4       183           84        Male       10  Married  Private         856.0   \n",
            "\n",
            "  Region  Annual.miles.drive  Years.noclaims  ...  Left.turn.intensity10  \\\n",
            "0  Urban             6213.71              25  ...                    1.0   \n",
            "1  Urban            12427.42              20  ...                   58.0   \n",
            "2  Urban            12427.42              14  ...                    0.0   \n",
            "3  Urban             6213.71              43  ...                    0.0   \n",
            "4  Urban             6213.71              65  ...                    2.0   \n",
            "\n",
            "   Left.turn.intensity11  Left.turn.intensity12  Right.turn.intensity08  \\\n",
            "0                    0.0                    0.0                     3.0   \n",
            "1                   24.0                   11.0                  1099.0   \n",
            "2                    0.0                    0.0                     0.0   \n",
            "3                    0.0                    0.0                     0.0   \n",
            "4                    0.0                    0.0                   325.0   \n",
            "\n",
            "   Right.turn.intensity09  Right.turn.intensity10  Right.turn.intensity11  \\\n",
            "0                     1.0                     0.0                     0.0   \n",
            "1                   615.0                   219.0                   101.0   \n",
            "2                     0.0                     0.0                     0.0   \n",
            "3                     0.0                     0.0                     0.0   \n",
            "4                   111.0                    18.0                     4.0   \n",
            "\n",
            "   Right.turn.intensity12  NB_Claim    AMT_Claim  \n",
            "0                     0.0         1  5100.171753  \n",
            "1                    40.0         1   883.554840  \n",
            "2                     0.0         0     0.000000  \n",
            "3                     0.0         0     0.000000  \n",
            "4                     2.0         0     0.000000  \n",
            "\n",
            "[5 rows x 52 columns]\n"
          ]
        }
      ],
      "source": [
        "# Mount Google Drive\n",
        "from google.colab import drive\n",
        "\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "# Specify file path\n",
        "\n",
        "file_path = '/content/drive/My Drive/telematics_syn.csv'\n",
        "\n",
        "# Import pandas (assuming you want to use it to read the CSV)\n",
        "import pandas as pd\n",
        "\n",
        "# Read the CSV file\n",
        "df = pd.read_csv(file_path)\n",
        "print(df.shape)  # Should print (100000, 52)\n",
        "print(df.head()) # To check the first few rows"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install pytorch-tabnet matplotlib scikit-learn"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-24rhrCUPBt_",
        "outputId": "b7a9dbf0-86ec-4c34-e342-f02b98e67f81"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: pytorch-tabnet in /usr/local/lib/python3.10/dist-packages (4.1.0)\n",
            "Requirement already satisfied: matplotlib in /usr/local/lib/python3.10/dist-packages (3.7.1)\n",
            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.10/dist-packages (1.5.2)\n",
            "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from pytorch-tabnet) (1.26.4)\n",
            "Requirement already satisfied: scipy>1.4 in /usr/local/lib/python3.10/dist-packages (from pytorch-tabnet) (1.13.1)\n",
            "Requirement already satisfied: torch>=1.3 in /usr/local/lib/python3.10/dist-packages (from pytorch-tabnet) (2.5.0+cu121)\n",
            "Requirement already satisfied: tqdm>=4.36 in /usr/local/lib/python3.10/dist-packages (from pytorch-tabnet) (4.66.5)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.3.0)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (0.12.1)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (4.54.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.4.7)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (24.1)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (10.4.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (3.2.0)\n",
            "Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (2.8.2)\n",
            "Requirement already satisfied: joblib>=1.2.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.4.2)\n",
            "Requirement already satisfied: threadpoolctl>=3.1.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (3.5.0)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch>=1.3->pytorch-tabnet) (3.16.1)\n",
            "Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.10/dist-packages (from torch>=1.3->pytorch-tabnet) (4.12.2)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch>=1.3->pytorch-tabnet) (3.4.2)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch>=1.3->pytorch-tabnet) (3.1.4)\n",
            "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch>=1.3->pytorch-tabnet) (2024.6.1)\n",
            "Requirement already satisfied: sympy==1.13.1 in /usr/local/lib/python3.10/dist-packages (from torch>=1.3->pytorch-tabnet) (1.13.1)\n",
            "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from sympy==1.13.1->torch>=1.3->pytorch-tabnet) (1.3.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch>=1.3->pytorch-tabnet) (3.0.2)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Import necessary libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from imblearn.over_sampling import SMOTE\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler, LabelEncoder, OneHotEncoder\n",
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.metrics import accuracy_score, f1_score\n",
        "from pytorch_tabnet.tab_model import TabNetClassifier\n",
        "import torch\n",
        "\n",
        "# Create a binary Risk Category label based on specified conditions\n",
        "df['Risk_Category'] = np.where((df['NB_Claim'] >= 1) & (df['AMT_Claim'] > 1000), 1, 0)\n",
        "\n",
        "# Drop NB_Claim and AMT_Claim as they are labels and should not be part of features\n",
        "df = df.drop(columns=['NB_Claim', 'AMT_Claim'])\n",
        "\n",
        "# Check the distribution of the new target variable\n",
        "print(df['Risk_Category'].value_counts())\n",
        "\n",
        "# Identify categorical columns and apply label encoding\n",
        "categorical_cols = df.select_dtypes(include=['object']).columns\n",
        "label_encoders = {}\n",
        "for col in categorical_cols:\n",
        "    le = LabelEncoder()\n",
        "    df[col] = le.fit_transform(df[col])\n",
        "    label_encoders[col] = le\n",
        "\n",
        "# Separate features and target variable\n",
        "X = df.drop('Risk_Category', axis=1)\n",
        "y = df['Risk_Category']\n",
        "\n",
        "# Apply SMOTE to balance the dataset\n",
        "smote = SMOTE(random_state=42)\n",
        "X_resampled, y_resampled = smote.fit_resample(X, y)\n",
        "\n",
        "# Convert resampled data to a DataFrame\n",
        "balanced_data = pd.concat([pd.DataFrame(X_resampled, columns=X.columns), pd.DataFrame(y_resampled, columns=['Risk_Category'])], axis=1)\n",
        "\n",
        "# Identify categorical and numerical columns\n",
        "categorical_cols = ['Insured.sex', 'Marital', 'Car.use', 'Region', 'Territory']\n",
        "numerical_cols = [col for col in balanced_data.columns if col not in categorical_cols + ['Risk_Category']]\n",
        "\n",
        "# Preprocess the data: one-hot encode categorical variables and normalize numerical features\n",
        "preprocessor = ColumnTransformer(\n",
        "    transformers=[\n",
        "        ('num', StandardScaler(), numerical_cols),\n",
        "        ('cat', OneHotEncoder(handle_unknown='ignore'), categorical_cols)\n",
        "    ])\n",
        "\n",
        "# Apply preprocessing\n",
        "X_preprocessed = preprocessor.fit_transform(balanced_data.drop('Risk_Category', axis=1))\n",
        "\n",
        "# Split the data into training, validation, and testing sets (60% train, 20% validation, 20% test)\n",
        "X_train, X_temp, y_train, y_temp = train_test_split(X_preprocessed, balanced_data['Risk_Category'], test_size=0.4, random_state=42, stratify=balanced_data['Risk_Category'])\n",
        "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42, stratify=y_temp)\n",
        "\n",
        "# Initialize the TabNet model with reduced epochs and learning rate to prevent overfitting\n",
        "tabnet_model = TabNetClassifier(\n",
        "    n_d=32, n_a=32, n_steps=5,\n",
        "    gamma=1.5, n_independent=2, n_shared=2,\n",
        "    cat_idxs=[], cat_dims=[], cat_emb_dim=[],\n",
        "    lambda_sparse=1e-3, optimizer_fn=torch.optim.Adam,\n",
        "    optimizer_params=dict(lr=1e-3),\n",
        "    mask_type=\"entmax\",\n",
        "    scheduler_params=dict(mode=\"min\", patience=5, min_lr=1e-5, factor=0.9),\n",
        "    scheduler_fn=torch.optim.lr_scheduler.ReduceLROnPlateau,\n",
        "    seed=42,\n",
        "    verbose=10\n",
        ")\n",
        "\n",
        "# Train the model with early stopping\n",
        "tabnet_model.fit(\n",
        "    X_train=X_train, y_train=y_train,\n",
        "    eval_set=[(X_train, y_train), (X_val, y_val)],\n",
        "    eval_name=['train', 'valid'],\n",
        "    eval_metric=['accuracy'],\n",
        "    max_epochs=10,  # Reduced to prevent overfitting\n",
        "    patience=5,\n",
        "    batch_size=1024,\n",
        "    virtual_batch_size=128,\n",
        "    num_workers=0,\n",
        "    drop_last=False\n",
        ")\n",
        "\n",
        "# Evaluate the model on the test set\n",
        "y_pred = tabnet_model.predict(X_test)\n",
        "\n",
        "# Calculate accuracy and F1 score\n",
        "accuracy = accuracy_score(y_test, y_pred)\n",
        "f1 = f1_score(y_test, y_pred, average='weighted')\n",
        "\n",
        "print(f\"Test Accuracy: {accuracy:.4f}\")\n",
        "print(f\"Test F1 Score: {f1:.4f}\")\n",
        "\n",
        "# Feature importance\n",
        "feature_importances = tabnet_model.feature_importances_\n",
        "feature_names = preprocessor.get_feature_names_out()\n",
        "importance_df = pd.DataFrame({'feature': feature_names, 'importance': feature_importances})\n",
        "importance_df = importance_df.sort_values('importance', ascending=False).head(10)\n",
        "print(\"Top 10 Important Features:\")\n",
        "print(importance_df)\n",
        "\n",
        "# Save the model\n",
        "saving_path_name = \"tabnet_model\"\n",
        "saved_filepath = tabnet_model.save_model(saving_path_name)\n",
        "print(f\"Model saved to: {saved_filepath}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "17T6sJlWGeQL",
        "outputId": "04c50152-8c30-454a-b5f0-3ee316e38cae"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Risk_Category\n",
            "0    97302\n",
            "1     2698\n",
            "Name: count, dtype: int64\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/pytorch_tabnet/abstract_model.py:82: UserWarning: Device used : cuda\n",
            "  warnings.warn(f\"Device used : {self.device}\")\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "epoch 0  | loss: 0.89424 | train_accuracy: 0.62311 | valid_accuracy: 0.62599 |  0:00:28s\n",
            "Stop training because you reached max_epochs = 10 with best_epoch = 9 and best_valid_accuracy = 0.75846\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/pytorch_tabnet/callbacks.py:172: UserWarning: Best weights from best epoch are automatically used!\n",
            "  warnings.warn(wrn_msg)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test Accuracy: 0.7544\n",
            "Test F1 Score: 0.7525\n",
            "Top 10 Important Features:\n",
            "                    feature  importance\n",
            "6    num__Annual.pct.driven    0.199205\n",
            "51           cat__Car.use_2    0.128139\n",
            "3         num__Credit.score    0.121263\n",
            "0             num__Duration    0.024281\n",
            "102       cat__Territory_58    0.021519\n",
            "77        cat__Territory_33    0.018302\n",
            "135       cat__Territory_91    0.015445\n",
            "125       cat__Territory_81    0.014457\n",
            "55        cat__Territory_11    0.014075\n",
            "68        cat__Territory_24    0.012182\n",
            "Successfully saved model at tabnet_model.zip\n",
            "Model saved to: tabnet_model.zip\n"
          ]
        }
      ]
    }
  ]
}