Boosting With lightgbm/Boosting_Implement_With_lightgbm.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "ONJmfkuS3rAR"
      },
      "outputs": [],
      "source": [
        "#*******************************************************************************************************************************************\n",
        "#                                                                                                                                          *\n",
        "#                                                Model Ensembles  -  Boosting                                                              *\n",
        "#                                                                                                                                          *\n",
        "#     - Implementation of a LightGBM model for binary classification on the Titanic dataset.                                               *\n",
        "#     - This code defines a LightGBMModel class, which initializes a LightGBM classifier with                                              *\n",
        "#       specified parameters such as the number of estimators and maximum depth.                                                           *\n",
        "#     - The class includes methods for fitting the model to training data, making predictions, and evaluating the model's performance.     *\n",
        "#     - It also provides visualization of the ROC curve and confusion matrix for evaluation.                                               *\n",
        "#     - The Titanic dataset is loaded, preprocessed, and split into training and testing sets.                                             *\n",
        "#     - Finally, the LightGBM model is initialized, trained on the training data, evaluated on the testing data,                           *\n",
        "#       and conclusions are drawn based on the evaluation.                                                                                 *\n",
        "#                                                                                                                                          *\n",
        "#*******************************************************************************************************************************************"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "x1BSGKkj305I"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import lightgbm as lgb\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report, roc_curve, confusion_matrix"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "id": "H7w6dTWm30vK"
      },
      "outputs": [],
      "source": [
        "class LightGBMModel:\n",
        "    def __init__(self, n_estimators=50, max_depth=1):\n",
        "        '''\n",
        "           Initialize LightGBM model with specified number of estimators and max depth\n",
        "        '''\n",
        "        self.n_estimators = n_estimators\n",
        "        self.max_depth = max_depth\n",
        "        self.model = lgb.LGBMClassifier(n_estimators=self.n_estimators, max_depth=self.max_depth)\n",
        "\n",
        "    def fit(self, X, y):\n",
        "        '''\n",
        "           Fit the LightGBM model to the training data\n",
        "        '''\n",
        "        self.model.fit(X, y)\n",
        "\n",
        "    def predict(self, X):\n",
        "        '''\n",
        "           Predict the class labels for the input data\n",
        "        '''\n",
        "        return self.model.predict(X)\n",
        "\n",
        "    def predict_proba(self, X):\n",
        "        '''\n",
        "           Predict class probabilities for the input data\n",
        "        '''\n",
        "        return self.model.predict_proba(X)[:, 1]\n",
        "\n",
        "    def evaluate_and_conclude(self, X_test, y_test):\n",
        "        '''\n",
        "           Evaluate the LightGBM model on the test data\n",
        "        '''\n",
        "        y_pred = self.predict(X_test)\n",
        "        print(\"*******************************************************************************************************************\")\n",
        "        print(\"Classification Report:\")\n",
        "        print(classification_report(y_test, y_pred))\n",
        "\n",
        "        print(\"*******************************************************************************************************************\\n\\n\")\n",
        "        y_pred_prob = self.predict_proba(X_test)\n",
        "        fpr, tpr, thresholds = roc_curve(y_test, y_pred_prob)\n",
        "        plt.figure(figsize=(8, 6))\n",
        "        plt.plot(fpr, tpr, label='ROC curve')\n",
        "        plt.plot([0, 1], [0, 1], 'k--', label='Random guess')\n",
        "        plt.xlabel('False Positive Rate')\n",
        "        plt.ylabel('True Positive Rate')\n",
        "        plt.title('ROC Curve')\n",
        "        plt.legend()\n",
        "        plt.show()\n",
        "        print(\"\\n\\n\")\n",
        "\n",
        "        conf_matrix = confusion_matrix(y_test, y_pred)\n",
        "        plt.figure(figsize=(8, 6))\n",
        "        plt.imshow(conf_matrix, cmap='Blues', interpolation='nearest')\n",
        "        plt.colorbar()\n",
        "        plt.title('Confusion Matrix')\n",
        "        plt.xlabel('Predicted Label')\n",
        "        plt.ylabel('True Label')\n",
        "        plt.xticks([0, 1], ['Not Survived', 'Survived'])\n",
        "        plt.yticks([0, 1], ['Not Survived', 'Survived'])\n",
        "        for i in range(2):\n",
        "            for j in range(2):\n",
        "                plt.text(j, i, str(conf_matrix[i, j]), ha='center', va='center', color='white')\n",
        "        plt.show()\n",
        "\n",
        "        self.banner()\n",
        "\n",
        "    def banner(self):\n",
        "        print(\"\"\"\n",
        "______________________________________________________________________________________________\n",
        "|                                                                                             |\n",
        "|         Conclusion and Suggestions:                                                         |\n",
        "|             Based on the evaluation of the LightGBM model and comparison with other models, |\n",
        "|             we observe a noticeable improvement in the classification results,              |\n",
        "|             indicating the effectiveness of the implemented algorithm.                      |\n",
        "|             This model speaks for itself in terms of performance.                           |\n",
        "|                                                                                             |\n",
        "|_____________________________________________________________________________________________|\n",
        "        \"\"\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Nn2TmDwU30gY",
        "outputId": "6f2562fc-49df-4717-93db-83e098430961"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "[LightGBM] [Info] Number of positive: 100, number of negative: 46\n",
            "[LightGBM] [Info] Auto-choosing row-wise multi-threading, the overhead of testing was 0.000236 seconds.\n",
            "You can set `force_row_wise=true` to remove the overhead.\n",
            "And if memory is not enough, you can set `force_col_wise=true`.\n",
            "[LightGBM] [Info] Total Bins 92\n",
            "[LightGBM] [Info] Number of data points in the train set: 146, number of used features: 7\n",
            "[LightGBM] [Info] [binary:BoostFromScore]: pavg=0.684932 -> initscore=0.776529\n",
            "[LightGBM] [Info] Start training from score 0.776529\n",
            "*******************************************************************************************************************\n",
            "Classification Report:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.75      0.64      0.69        14\n",
            "           1       0.80      0.87      0.83        23\n",
            "\n",
            "    accuracy                           0.78        37\n",
            "   macro avg       0.78      0.76      0.76        37\n",
            "weighted avg       0.78      0.78      0.78        37\n",
            "\n",
            "*******************************************************************************************************************\n",
            "\n",
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x600 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x600 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "______________________________________________________________________________________________\n",
            "|                                                                                             |\n",
            "|         Conclusion and Suggestions:                                                         |\n",
            "|             Based on the evaluation of the LightGBM model and comparison with other models, |\n",
            "|             we observe a noticeable improvement in the classification results,              |\n",
            "|             indicating the effectiveness of the implemented algorithm.                      |\n",
            "|             This model speaks for itself in terms of performance.                           |\n",
            "|                                                                                             |\n",
            "|_____________________________________________________________________________________________|\n",
            "        \n"
          ]
        }
      ],
      "source": [
        "titanic_df = pd.read_csv(\"titanic.csv\")                                                         # Load the Titanic dataset\n",
        "titanic_df = titanic_df.dropna()                                                                # Preprocess the data (you may need to handle missing values, encode categorical variables, etc.)\n",
        "X = titanic_df[['Pclass', 'Age', 'Fare', 'Sex', 'SibSp', 'Parch', 'Embarked']].values\n",
        "y = titanic_df['Survived'].values\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)       # Split the data into train and test sets\n",
        "label_encoder = LabelEncoder()\n",
        "X_train[:, 3] = label_encoder.fit_transform(X_train[:, 3])\n",
        "X_test[:, 3] = label_encoder.transform(X_test[:, 3])\n",
        "label_encoder = LabelEncoder()\n",
        "X_train[:, 6] = label_encoder.fit_transform(X_train[:, 6])\n",
        "X_test[:, 6] = label_encoder.transform(X_test[:, 6])\n",
        "clf = LightGBMModel(n_estimators=50, max_depth=1)                                               # Initialize and train LightGBM model\n",
        "clf.fit(X_train, y_train)\n",
        "clf.evaluate_and_conclude(X_test, y_test)                                                       # Evaluate and conclude"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}