{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.preprocessing import StandardScaler, OneHotEncoder\n",
        "from sklearn.compose import ColumnTransformer\n",
        "from sklearn.pipeline import Pipeline\n",
        "import joblib"
      ],
      "metadata": {
        "id": "a1U4QR1AOKAK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load processed data\n",
        "train_df = pd.read_csv('train_processed.csv')\n",
        "test_df = pd.read_csv('test_processed.csv')"
      ],
      "metadata": {
        "id": "mOjH5CqhOlTp"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Prepare features and target\n",
        "X_train = train_df.drop(['is_fraud', 'trans_date_trans_time', 'trans_num', 'dob', 'unix_time'], axis=1)\n",
        "y_train = train_df['is_fraud']\n",
        "\n",
        "X_test = test_df.drop(['is_fraud', 'trans_date_trans_time', 'trans_num', 'dob', 'unix_time'], axis=1)\n",
        "y_test = test_df['is_fraud']\n"
      ],
      "metadata": {
        "id": "VW8cpQ6FNvC9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Define preprocessing\n",
        "numeric_features = ['amt', 'city_pop', 'hour', 'day_of_week', 'month', 'age', 'distance']\n",
        "numeric_transformer = StandardScaler()\n",
        "\n",
        "categorical_features = ['category', 'gender', 'job', 'merchant']\n",
        "categorical_transformer = OneHotEncoder(handle_unknown='ignore')\n",
        "\n",
        "preprocessor = ColumnTransformer(\n",
        "    transformers=[\n",
        "        ('num', numeric_transformer, numeric_features),\n",
        "        ('cat', categorical_transformer, categorical_features)\n",
        "    ])"
      ],
      "metadata": {
        "id": "GkGMEEYqOc_p"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Create pipeline\n",
        "model = Pipeline(steps=[\n",
        "    ('preprocessor', preprocessor),\n",
        "    ('classifier', RandomForestClassifier(class_weight='balanced', random_state=42))\n",
        "])\n"
      ],
      "metadata": {
        "id": "s2VyUaV6OZw9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train model\n",
        "model.fit(X_train, y_train)"
      ],
      "metadata": {
        "id": "6dXK4nJnOW40"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate on test data\n",
        "y_pred = model.predict(X_test)\n",
        "print(\"Test Set Performance:\")\n",
        "print(classification_report(y_test, y_pred))"
      ],
      "metadata": {
        "id": "PNV9fZhkOVFM"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Save model\n",
        "joblib.dump(model, 'fraud_model.pkl')\n",
        "print(\"Model saved to fraud_model.pkl\")"
      ],
      "metadata": {
        "id": "4j2mlzRINc_N"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}