{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "d0fc441d-a873-4a73-8afb-5643f99ca4c3", "metadata": { "scrolled": true }, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "from sklearn.model_selection import train_test_split, GridSearchCV\n", "from sklearn.preprocessing import StandardScaler\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.metrics import classification_report\n", "from sklearn.feature_selection import SelectKBest, f_classif\n", "import warnings\n" ] }, { "cell_type": "code", "execution_count": 14, "id": "d995d276-13dc-4cc1-83e9-d89295ad2d08", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--- Step 1: Dataset Loading and Cleaning ---\n", "Dataset loaded successfully.\n", "Cleaned Column Names:\n", "Index(['age', 'sex', 'cp', 'trestbps', 'chol', 'fbs', 'restecg', 'thalach',\n", " 'exang', 'oldpeak', 'slope', 'ca', 'thal', 'num'],\n", " dtype='object')\n" ] } ], "source": [ "print(\"--- Step 1: Dataset Loading and Cleaning ---\")\n", "try:\n", " df = pd.read_csv('Heart Attack Prediction.csv')\n", " df.columns = df.columns.str.strip()\n", " print(\"Dataset loaded successfully.\")\n", " print(\"Cleaned Column Names:\")\n", " print(df.columns)\n", "except FileNotFoundError:\n", " print(\"Error: 'Heart Attack Prediction.csv' file not found.\")\n", " exit()" ] }, { "cell_type": "code", "execution_count": 15, "id": "937e6615-6661-4dff-ab16-553806b16115", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--- Step 2: Handling Missing Values ---\n", "Missing values handled.\n", "Missing values after imputation:\n", "age 0\n", "sex 0\n", "cp 0\n", "trestbps 0\n", "chol 0\n", "fbs 0\n", "restecg 0\n", "thalach 0\n", "exang 0\n", "oldpeak 0\n", "slope 0\n", "ca 0\n", "thal 0\n", "num 0\n", "dtype: int64\n" ] } ], "source": [ "# Step 2: Handle Missing Values\n", "print(\"--- Step 2: Handling Missing Values ---\")\n", "df['ca'] = df['ca'].replace('?', np.nan)\n", "df['thal'] = df['thal'].replace('?', np.nan)\n", "df['ca'] = pd.to_numeric(df['ca'])\n", "df['thal'] = pd.to_numeric(df['thal'])\n", "df['ca'] = df['ca'].replace(4, np.nan)\n", "df['thal'] = df['thal'].replace(0, np.nan)\n", "for col in ['ca', 'thal']:\n", " df[col] = df[col].fillna(df[col].mode()[0])\n", "print(\"Missing values handled.\")\n", "print(\"Missing values after imputation:\")\n", "print(df.isnull().sum())" ] }, { "cell_type": "code", "execution_count": 16, "id": "b366e7c8-9ecb-456b-910c-d60e81df4982", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--- Step 3: Applying ML Models with Grid Search ---\n", "Data split and scaled.\n" ] } ], "source": [ "# Step 3: Apply ML Models with Grid Search\n", "print(\"--- Step 3: Applying ML Models with Grid Search ---\")\n", "target_column_name = \"num\"\n", "try:\n", " X = df.drop(target_column_name, axis=1)\n", " y = df[target_column_name]\n", "except KeyError:\n", " print(f\"Error: The target column '{target_column_name}' was not found. Please re-check the column names.\")\n", " exit()\n", "\n", "for col in X.columns:\n", " X[col] = pd.to_numeric(X[col], errors='coerce')\n", "X = X.dropna(axis=1, how='any')\n", "X = X.dropna()\n", "y = y[X.index]\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "scaler = StandardScaler()\n", "X_train_scaled = scaler.fit_transform(X_train)\n", "X_test_scaled = scaler.transform(X_test)\n", "print(\"Data split and scaled.\")" ] }, { "cell_type": "code", "execution_count": 17, "id": "01a809ff-d653-4728-bd1c-08cd58db8bac", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "--- Logistic Regression Grid Search ---\n", "Logistic Regression Grid Search completed.\n", "Best parameters: {'C': 0.001, 'solver': 'liblinear'}\n", "\n", "Classification Report on Test Data (Logistic Regression):\n", " precision recall f1-score support\n", "\n", " 0 0.85 0.76 0.81 38\n", " 1 0.64 0.76 0.70 21\n", "\n", " accuracy 0.76 59\n", " macro avg 0.75 0.76 0.75 59\n", "weighted avg 0.78 0.76 0.77 59\n", "\n" ] } ], "source": [ "# Logistic Regression\n", "print(\"\\n--- Logistic Regression Grid Search ---\")\n", "param_grid_lr = {'C': [0.001, 0.01, 0.1, 1, 10, 100], 'solver': ['liblinear']}\n", "grid_search_lr = GridSearchCV(LogisticRegression(max_iter=1000), param_grid_lr, cv=5)\n", "with warnings.catch_warnings():\n", " warnings.filterwarnings('ignore', category=UserWarning)\n", " grid_search_lr.fit(X_train_scaled, y_train)\n", "print(\"Logistic Regression Grid Search completed.\")\n", "print(\"Best parameters:\", grid_search_lr.best_params_)\n", "y_pred_lr = grid_search_lr.predict(X_test_scaled)\n", "print(\"\\nClassification Report on Test Data (Logistic Regression):\")\n", "print(classification_report(y_test, y_pred_lr))" ] }, { "cell_type": "code", "execution_count": 18, "id": "f6c2e4e0-4f5a-459d-955f-5acaaeafd5a0", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "--- Random Forest Grid Search ---\n", "Random Forest Grid Search completed.\n", "Best parameters: {'max_depth': 5, 'n_estimators': 200}\n", "\n", "Classification Report on Test Data (Random Forest):\n", " precision recall f1-score support\n", "\n", " 0 0.83 0.79 0.81 38\n", " 1 0.65 0.71 0.68 21\n", "\n", " accuracy 0.76 59\n", " macro avg 0.74 0.75 0.75 59\n", "weighted avg 0.77 0.76 0.76 59\n", "\n" ] } ], "source": [ "# Random Forest\n", "print(\"\\n--- Random Forest Grid Search ---\")\n", "param_grid_rf = {'n_estimators': [50, 100, 200], 'max_depth': [5, 10, 15]}\n", "grid_search_rf = GridSearchCV(RandomForestClassifier(random_state=42), param_grid_rf, cv=5)\n", "grid_search_rf.fit(X_train, y_train)\n", "print(\"Random Forest Grid Search completed.\")\n", "print(\"Best parameters:\", grid_search_rf.best_params_)\n", "y_pred_rf = grid_search_rf.predict(X_test)\n", "print(\"\\nClassification Report on Test Data (Random Forest):\")\n", "print(classification_report(y_test, y_pred_rf))" ] }, { "cell_type": "code", "execution_count": 19, "id": "1db39c9b-9431-4be8-9b4e-38fb150d9c8d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--- Step 4: Performing Feature Selection ---\n", "Feature importance from Random Forest calculated.\n", "\n", "Feature Importance from Random Forest:\n", " feature importance\n", "3 oldpeak 0.389775\n", "2 cp 0.352867\n", "0 age 0.139262\n", "1 sex 0.090855\n", "5 thal 0.027241\n", "4 ca 0.000000\n", "\n", "Warning: The following features have constant values and will be removed: ['ca']\n", "\n", "Top 5 features selected by SelectKBest.\n", "Top 5 Features selected by SelectKBest:\n", "['age', 'sex', 'cp', 'oldpeak', 'thal']\n" ] } ], "source": [ "# Step 4: Perform Feature Selection\n", "print(\"--- Step 4: Performing Feature Selection ---\")\n", "importances = grid_search_rf.best_estimator_.feature_importances_\n", "feature_names = X.columns\n", "feature_importance_df = pd.DataFrame({'feature': feature_names, 'importance': importances}).sort_values(by='importance', ascending=False)\n", "print(\"Feature importance from Random Forest calculated.\")\n", "print(\"\\nFeature Importance from Random Forest:\")\n", "print(feature_importance_df)\n", "k = 5\n", "constant_features = [col for col in X.columns if X[col].nunique() == 1]\n", "if constant_features:\n", " print(f\"\\nWarning: The following features have constant values and will be removed: {constant_features}\")\n", " X_filtered = X.drop(columns=constant_features)\n", "else:\n", " X_filtered = X.copy()\n", "selector = SelectKBest(f_classif, k=k)\n", "selector.fit(X_filtered, y)\n", "selected_features_mask = selector.get_support()\n", "selected_features_kbest = X_filtered.columns[selected_features_mask]\n", "print(f\"\\nTop {k} features selected by SelectKBest.\")\n", "print(f\"Top {k} Features selected by SelectKBest:\")\n", "print(selected_features_kbest.tolist())" ] }, { "cell_type": "code", "execution_count": 20, "id": "100cf5b5-59e9-4f5e-8ff3-531097f68a2b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--- Step 5: Applying Models to Selected Features ---\n", "Data split and scaled with selected features.\n" ] } ], "source": [ "# Step 5: Apply Models to Selected Features\n", "print(\"--- Step 5: Applying Models to Selected Features ---\")\n", "X_selected = X[selected_features_kbest]\n", "X_train_sel, X_test_sel, y_train_sel, y_test_sel = train_test_split(X_selected, y, test_size=0.2, random_state=42)\n", "scaler_sel = StandardScaler()\n", "X_train_sel_scaled = scaler_sel.fit_transform(X_train_sel)\n", "X_test_sel_scaled = scaler_sel.transform(X_test_sel)\n", "print(\"Data split and scaled with selected features.\")" ] }, { "cell_type": "code", "execution_count": 21, "id": "e743baa6-eec2-40de-978b-c74fff2b7422", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "--- Logistic Regression with Selected Features ---\n", "Model trained and predictions made.\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " 0 0.85 0.76 0.81 38\n", " 1 0.64 0.76 0.70 21\n", "\n", " accuracy 0.76 59\n", " macro avg 0.75 0.76 0.75 59\n", "weighted avg 0.78 0.76 0.77 59\n", "\n" ] } ], "source": [ "# Logistic Regression with selected features\n", "lr_selected = LogisticRegression(**grid_search_lr.best_params_)\n", "lr_selected.fit(X_train_sel_scaled, y_train_sel)\n", "y_pred_lr_sel = lr_selected.predict(X_test_sel_scaled)\n", "print(\"\\n--- Logistic Regression with Selected Features ---\")\n", "print(\"Model trained and predictions made.\")\n", "print(\"Classification Report:\")\n", "print(classification_report(y_test_sel, y_pred_lr_sel))" ] }, { "cell_type": "code", "execution_count": 22, "id": "ee99ce5b-1f6f-4fae-ac2c-0c0939e4d483", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "--- Random Forest with Selected Features ---\n", "Model trained and predictions made.\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " 0 0.84 0.82 0.83 38\n", " 1 0.68 0.71 0.70 21\n", "\n", " accuracy 0.78 59\n", " macro avg 0.76 0.77 0.76 59\n", "weighted avg 0.78 0.78 0.78 59\n", "\n" ] } ], "source": [ "# Random Forest with selected features\n", "rf_selected = RandomForestClassifier(**grid_search_rf.best_params_, random_state=42)\n", "rf_selected.fit(X_train_sel, y_train_sel)\n", "y_pred_rf_sel = rf_selected.predict(X_test_sel)\n", "print(\"\\n--- Random Forest with Selected Features ---\")\n", "print(\"Model trained and predictions made.\")\n", "print(\"Classification Report:\")\n", "print(classification_report(y_test_sel, y_pred_rf_sel))" ] }, { "cell_type": "code", "execution_count": 23, "id": "7acd8b1b-923e-431b-a368-8f682563473d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "--- Step 6: Saving and Final Review ---\n", "Processed dataset saved to: Heart_Attack_Processed_Dataset.csv\n", "\n", "Final DataFrame Info after all operations:\n", "\n", "RangeIndex: 294 entries, 0 to 293\n", "Data columns (total 14 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 age 294 non-null int64 \n", " 1 sex 294 non-null int64 \n", " 2 cp 294 non-null int64 \n", " 3 trestbps 294 non-null object \n", " 4 chol 294 non-null object \n", " 5 fbs 294 non-null object \n", " 6 restecg 294 non-null object \n", " 7 thalach 294 non-null object \n", " 8 exang 294 non-null object \n", " 9 oldpeak 294 non-null float64\n", " 10 slope 294 non-null object \n", " 11 ca 294 non-null float64\n", " 12 thal 294 non-null float64\n", " 13 num 294 non-null int64 \n", "dtypes: float64(3), int64(4), object(7)\n", "memory usage: 32.3+ KB\n", "\n", "First 5 rows of the processed DataFrame:\n", " age sex cp trestbps chol fbs restecg thalach exang oldpeak slope ca \\\n", "0 28 1 2 130 132 0 2 185 0 0.0 ? 0.0 \n", "1 29 1 2 120 243 0 0 160 0 0.0 ? 0.0 \n", "2 29 1 2 140 ? 0 0 170 0 0.0 ? 0.0 \n", "3 30 0 1 170 237 0 1 170 0 0.0 ? 0.0 \n", "4 31 0 2 100 219 0 1 150 0 0.0 ? 0.0 \n", "\n", " thal num \n", "0 7.0 0 \n", "1 7.0 0 \n", "2 7.0 0 \n", "3 6.0 0 \n", "4 7.0 0 \n" ] } ], "source": [ "# Step 6: Final Output\n", "print(\"--- Step 6: Saving and Final Review ---\")\n", "output_file_name = 'Heart_Attack_Processed_Dataset.csv'\n", "df.to_csv(output_file_name, index=False)\n", "print(f\"Processed dataset saved to: {output_file_name}\")\n", "\n", "print(\"\\nFinal DataFrame Info after all operations:\")\n", "df.info()\n", "\n", "print(\"\\nFirst 5 rows of the processed DataFrame:\")\n", "print(df.head())" ] }, { "cell_type": "code", "execution_count": null, "id": "b726273a-a35d-4910-9f7b-49d445ce5700", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python [conda env:base] *", "language": "python", "name": "conda-base-py" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.7" } }, "nbformat": 4, "nbformat_minor": 5 }