{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "1a0c7251-b961-4f5f-b239-20465cb60faf", "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "from sklearn.model_selection import train_test_split,GridSearchCV\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.metrics import classification_report, confusion_matrix\n", "import seaborn as sns\n", "from sklearn.preprocessing import LabelEncoder\n", "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.metrics import accuracy_score\n", "from sklearn.model_selection import StratifiedKFold,cross_val_score" ] }, { "cell_type": "code", "execution_count": 3, "id": "4fb048df-9c0a-44d2-9087-0dc2f4c6b5cd", "metadata": {}, "outputs": [], "source": [ "data = pd.read_csv('Heart Disease.csv')" ] }, { "cell_type": "code", "execution_count": 7, "id": "60c9b2f2-2a0e-4b4f-a25d-1e8d4c3835b1", "metadata": {}, "outputs": [], "source": [ "df = pd.DataFrame(data)" ] }, { "cell_type": "code", "execution_count": 9, "id": "4321f33b-f37d-46ca-abbe-78f0c0a2cceb", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
agegenderimplusepressurehightpressurelowglucosekcmtroponinclass
06416616083160.01.800.012negative
1211949846296.06.751.060positive
25516416077270.01.990.003negative
36417012055270.013.870.122positive
45516411265300.01.080.003negative
..............................
13094719410581135.036.240.263positive
13107008013575351.02.2110.000positive
131185111211569114.02.190.062positive
1312481841186896.05.330.006negative
13138604017968147.05.220.011negative
\n", "

1314 rows × 9 columns

\n", "
" ], "text/plain": [ " age gender impluse pressurehight pressurelow glucose kcm \\\n", "0 64 1 66 160 83 160.0 1.80 \n", "1 21 1 94 98 46 296.0 6.75 \n", "2 55 1 64 160 77 270.0 1.99 \n", "3 64 1 70 120 55 270.0 13.87 \n", "4 55 1 64 112 65 300.0 1.08 \n", "... ... ... ... ... ... ... ... \n", "1309 47 1 94 105 81 135.0 36.24 \n", "1310 70 0 80 135 75 351.0 2.21 \n", "1311 85 1 112 115 69 114.0 2.19 \n", "1312 48 1 84 118 68 96.0 5.33 \n", "1313 86 0 40 179 68 147.0 5.22 \n", "\n", " troponin class \n", "0 0.012 negative \n", "1 1.060 positive \n", "2 0.003 negative \n", "3 0.122 positive \n", "4 0.003 negative \n", "... ... ... \n", "1309 0.263 positive \n", "1310 10.000 positive \n", "1311 0.062 positive \n", "1312 0.006 negative \n", "1313 0.011 negative \n", "\n", "[1314 rows x 9 columns]" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.head(-5)" ] }, { "cell_type": "code", "execution_count": 11, "id": "d7e26349-68bf-4e93-afc3-e43f9bf07ba5", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "age 0\n", "gender 0\n", "impluse 0\n", "pressurehight 0\n", "pressurelow 0\n", "glucose 0\n", "kcm 0\n", "troponin 0\n", "class 0\n", "dtype: int64" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.isna().sum()" ] }, { "cell_type": "code", "execution_count": 13, "id": "ade1b7da-65c2-43cf-95c4-f31f72ecc0d3", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "age int64\n", "gender int64\n", "impluse int64\n", "pressurehight int64\n", "pressurelow int64\n", "glucose float64\n", "kcm float64\n", "troponin float64\n", "class object\n", "dtype: object" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.dtypes" ] }, { "cell_type": "code", "execution_count": 15, "id": "1182db84-02b0-469a-ae35-c4f777a4fcec", "metadata": {}, "outputs": [], "source": [ "class_counts = df['class'].value_counts() " ] }, { "cell_type": "code", "execution_count": 17, "id": "00abbaff-4b50-4b45-9b61-0ce8881a512b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ " age gender impluse pressurehight pressurelow glucose kcm troponin \\\n", "0 64 1 66 160 83 160.0 1.80 0.012 \n", "1 21 1 94 98 46 296.0 6.75 1.060 \n", "2 55 1 64 160 77 270.0 1.99 0.003 \n", "3 64 1 70 120 55 270.0 13.87 0.122 \n", "4 55 1 64 112 65 300.0 1.08 0.003 \n", "\n", " class \n", "0 negative \n", "1 positive \n", "2 negative \n", "3 positive \n", "4 negative \n" ] } ], "source": [ "print(df.head())" ] }, { "cell_type": "code", "execution_count": 19, "id": "458542c6-873c-4566-b5e2-e36b66f00981", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "class\n", "positive 810\n", "negative 509\n", "Name: count, dtype: int64" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" } ], "source": [ "data['class'].value_counts()" ] }, { "cell_type": "code", "execution_count": 21, "id": "960bb3b0-3d71-4602-b2d9-c077ccec3055", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Feature Shape: (1319, 8)\n", "Target Shape: (1319,)\n" ] } ], "source": [ "# Define features (X) and target (y)\n", "X = data.drop(\"class\", axis=1) # All columns except \"class\"\n", "y = (data[\"class\"] == \"positive\").astype(int) # Convert target to binary (0: negative, 1: positive)\n", "\n", "# Print shape of data\n", "print(\"Feature Shape:\", X.shape)\n", "print(\"Target Shape:\", y.shape)" ] }, { "cell_type": "code", "execution_count": 23, "id": "9a880740-2f9f-4447-846a-f2720ecaf76a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Training Data Shape: (1055, 8)\n", "Testing Data Shape: (264, 8)\n" ] } ], "source": [ "from sklearn.model_selection import train_test_split\n", "\n", "# Define features (X) and target (y)\n", "X = data.drop(\"class\", axis=1) # Features\n", "y = (data[\"class\"] == \"positive\").astype(int) # Convert labels to binary\n", "\n", "# Split data (80% train, 20% test)\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Print shapes\n", "print(\"Training Data Shape:\", X_train.shape)\n", "print(\"Testing Data Shape:\", X_test.shape)" ] }, { "cell_type": "code", "execution_count": 27, "id": "1ca33902-c1fe-472c-819d-43ed040bf552", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.80\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " negative 0.75 0.69 0.72 101\n", " positive 0.82 0.86 0.84 163\n", "\n", " accuracy 0.80 264\n", " macro avg 0.79 0.78 0.78 264\n", "weighted avg 0.79 0.80 0.79 264\n", "\n" ] } ], "source": [ "from sklearn.linear_model import LogisticRegression\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import accuracy_score, classification_report\n", "import pandas as pd\n", "\n", "# Load dataset\n", "df = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Define features and target\n", "X = df.drop(columns=[\"class\"]) # Exclude target variable\n", "y = df[\"class\"]\n", "\n", "# Split data into training and testing sets\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Initialize and train Logistic Regression model\n", "lr = LogisticRegression(max_iter=1000)\n", "lr.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = lr.predict(X_test)\n", "\n", "# Evaluate model\n", "accuracy = accuracy_score(y_test, y_pred)\n", "print(f\"Accuracy: {accuracy:.2f}\")\n", "print(\"Classification Report:\\n\", classification_report(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": 29, "id": "5b0da180-3e10-4eb8-aff2-b98ff29f9b92", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.98\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " negative 0.97 0.97 0.97 101\n", " positive 0.98 0.98 0.98 163\n", "\n", " accuracy 0.98 264\n", " macro avg 0.98 0.98 0.98 264\n", "weighted avg 0.98 0.98 0.98 264\n", "\n" ] } ], "source": [ "from sklearn.tree import DecisionTreeClassifier\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import accuracy_score, classification_report\n", "import pandas as pd\n", "\n", "# Load dataset\n", "df = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Define features and target\n", "X = df.drop(columns=[\"class\"]) # Exclude the target column\n", "y = df[\"class\"]\n", "\n", "# Split data into training and testing sets\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Initialize and train Decision Tree model\n", "dt = DecisionTreeClassifier(max_depth=5, random_state=42)\n", "dt.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = dt.predict(X_test)\n", "\n", "# Evaluate model\n", "accuracy = accuracy_score(y_test, y_pred)\n", "print(f\"Accuracy: {accuracy:.2f}\")\n", "print(\"Classification Report:\\n\", classification_report(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": 31, "id": "1c56e6c6-5204-40d9-987c-46e3ed620306", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.64\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " negative 0.53 0.55 0.54 101\n", " positive 0.72 0.70 0.71 163\n", "\n", " accuracy 0.64 264\n", " macro avg 0.63 0.63 0.63 264\n", "weighted avg 0.65 0.64 0.65 264\n", "\n" ] } ], "source": [ "from sklearn.neighbors import KNeighborsClassifier\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import accuracy_score, classification_report\n", "import pandas as pd\n", "\n", "# Load dataset\n", "df = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Define features and target variable\n", "X = df.drop(columns=[\"class\"]) # Exclude target column\n", "y = df[\"class\"]\n", "\n", "# Split dataset into training and testing sets\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Initialize and train KNN model\n", "knn = KNeighborsClassifier(n_neighbors=5)\n", "knn.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = knn.predict(X_test)\n", "\n", "# Evaluate accuracy\n", "accuracy = accuracy_score(y_test, y_pred)\n", "print(f\"Accuracy: {accuracy:.2f}\")\n", "print(\"Classification Report:\\n\", classification_report(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": 33, "id": "72268b6f-6d90-4d62-ba92-aec24a48be09", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.69\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " negative 0.75 0.30 0.43 101\n", " positive 0.68 0.94 0.79 163\n", "\n", " accuracy 0.69 264\n", " macro avg 0.72 0.62 0.61 264\n", "weighted avg 0.71 0.69 0.65 264\n", "\n" ] } ], "source": [ "from sklearn.svm import SVC\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import accuracy_score, classification_report\n", "import pandas as pd\n", "\n", "# Load dataset\n", "df = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Define features and target variable\n", "X = df.drop(columns=[\"class\"]) # Exclude the target column\n", "y = df[\"class\"]\n", "\n", "# Split dataset into training and testing sets\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Initialize and train SVM model\n", "svm = SVC(kernel='rbf', C=1.0, gamma='scale') # RBF kernel for nonlinear classification\n", "svm.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = svm.predict(X_test)\n", "\n", "# Evaluate accuracy\n", "accuracy = accuracy_score(y_test, y_pred)\n", "print(f\"Accuracy: {accuracy:.2f}\")\n", "print(\"Classification Report:\\n\", classification_report(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": 35, "id": "4e6faa26-91a3-4afb-8ffa-0e698bdaf18b", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "from sklearn.metrics import confusion_matrix\n", "\n", "# Compute confusion matrix\n", "cm = confusion_matrix(y_test, y_pred)\n", "\n", "# Plot confusion matrix\n", "plt.figure(figsize=(6, 4))\n", "sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=[\"Negative\", \"Positive\"], yticklabels=[\"Negative\", \"Positive\"])\n", "plt.xlabel(\"Predicted Labels\")\n", "plt.ylabel(\"True Labels\")\n", "plt.title(\"Confusion Matrix\")\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 37, "id": "beed8ca5-a975-4396-828c-420bcd3acb93", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Best Parameters: {'max_depth': None, 'min_samples_split': 2, 'n_estimators': 100}\n" ] } ], "source": [ "from sklearn.model_selection import GridSearchCV\n", "from sklearn.ensemble import RandomForestClassifier\n", "import pandas as pd\n", "\n", "# Load dataset\n", "data = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Prepare features and target\n", "X = data.drop(\"class\", axis=1)\n", "y = (data[\"class\"] == \"positive\").astype(int)\n", "\n", "# Define model and hyperparameter grid\n", "rf = RandomForestClassifier()\n", "param_grid = {\n", " \"n_estimators\": [50, 100, 200],\n", " \"max_depth\": [None, 10, 20],\n", " \"min_samples_split\": [2, 5, 10]\n", "}\n", "\n", "# Apply Grid Search\n", "grid_search = GridSearchCV(rf, param_grid, cv=5, scoring=\"accuracy\")\n", "grid_search.fit(X, y)\n", "\n", "# Print best parameters\n", "print(\"Best Parameters:\", grid_search.best_params_)" ] }, { "cell_type": "code", "execution_count": 39, "id": "5daf87ee-1cea-42a8-aa0a-56f96ca1d3d4", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Cross-Validation Accuracy Scores: [0.97727273 0.98484848 0.99242424 0.98106061 0.98859316]\n", "Mean Accuracy: 0.9848398432999194\n" ] } ], "source": [ "from sklearn.model_selection import cross_val_score, KFold\n", "from sklearn.ensemble import RandomForestClassifier\n", "import pandas as pd\n", "\n", "# Load dataset\n", "data = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Prepare features and target\n", "X = data.drop(\"class\", axis=1)\n", "y = (data[\"class\"] == \"positive\").astype(int)\n", "\n", "# Define model\n", "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n", "\n", "# Apply K-Fold Cross-Validation\n", "kf = KFold(n_splits=5, shuffle=True, random_state=42)\n", "scores = cross_val_score(rf, X, y, cv=kf, scoring=\"accuracy\")\n", "\n", "# Print results\n", "print(\"Cross-Validation Accuracy Scores:\", scores)\n", "print(\"Mean Accuracy:\", scores.mean())" ] }, { "cell_type": "code", "execution_count": 41, "id": "a581e4df-4cf6-4e46-9e36-0b0c69c0879e", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Random Forest Accuracy: 0.9810606060606061\n" ] } ], "source": [ "from sklearn.ensemble import RandomForestClassifier\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.metrics import accuracy_score\n", "import pandas as pd\n", "\n", "# Load dataset\n", "data = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Prepare features and target\n", "X = data.drop(\"class\", axis=1)\n", "y = (data[\"class\"] == \"positive\").astype(int)\n", "\n", "# Split data\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Train Random Forest model\n", "rf = RandomForestClassifier(n_estimators=100, random_state=42)\n", "rf.fit(X_train, y_train)\n", "\n", "# Make predictions\n", "y_pred = rf.predict(X_test)\n", "\n", "# Evaluate model\n", "accuracy = accuracy_score(y_test, y_pred)\n", "print(\"Random Forest Accuracy:\", accuracy)" ] }, { "cell_type": "code", "execution_count": 43, "id": "21741133-6f75-48cf-b65a-73d8dc9abfa2", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Fitting 5 folds for each of 27 candidates, totalling 135 fits\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time= 0.0s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time= 0.3s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time= 0.8s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=None, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time= 0.2s\n", "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=10, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=10, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 0.8s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 0.8s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=10, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time= 0.2s\n", "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time= 0.2s\n", "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time= 0.2s\n", "[CV] END .max_depth=20, min_samples_split=2, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=2, n_estimators=200; total time= 0.9s\n", "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time= 0.1s\n", "[CV] END .max_depth=20, min_samples_split=5, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 0.8s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=5, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time= 0.2s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=50; total time= 0.1s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=100; total time= 0.4s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 0.9s\n", "[CV] END max_depth=20, min_samples_split=10, n_estimators=200; total time= 0.8s\n", "Best Parameters: {'max_depth': None, 'min_samples_split': 2, 'n_estimators': 200}\n" ] } ], "source": [ "from sklearn.model_selection import GridSearchCV\n", "from sklearn.ensemble import RandomForestClassifier\n", "\n", "# Define hyperparameter grid for Random Forest\n", "param_grid = {\n", " \"n_estimators\": [50, 100, 200],\n", " \"max_depth\": [None, 10, 20],\n", " \"min_samples_split\": [2, 5, 10]\n", "}\n", "\n", "# Initialize model\n", "rf = RandomForestClassifier()\n", "\n", "# Apply Grid Search with 5-fold Cross-Validation\n", "grid_search = GridSearchCV(rf, param_grid, cv=5, scoring=\"accuracy\", verbose=2)\n", "grid_search.fit(X_train, y_train)\n", "\n", "# Print best parameters\n", "print(\"Best Parameters:\", grid_search.best_params_)" ] }, { "cell_type": "code", "execution_count": 45, "id": "a701626b-ec57-4f28-8145-201b1059ff48", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Cross-Validation Accuracy Scores: [0.98104265 0.98104265 0.98578199 0.99526066 0.99052133]\n", "Mean Accuracy: 0.9867298578199051\n" ] } ], "source": [ "from sklearn.model_selection import cross_val_score\n", "\n", "# Perform K-Fold Cross-Validation\n", "scores = cross_val_score(grid_search.best_estimator_, X_train, y_train, cv=5, scoring=\"accuracy\")\n", "\n", "# Print results\n", "print(\"Cross-Validation Accuracy Scores:\", scores)\n", "print(\"Mean Accuracy:\", scores.mean())" ] }, { "cell_type": "code", "execution_count": 47, "id": "26681905-1375-4cfd-96b6-b68338186ccf", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Final Model Accuracy: 0.9772727272727273\n", "Classification Report:\n", " precision recall f1-score support\n", "\n", " 0 0.97 0.97 0.97 101\n", " 1 0.98 0.98 0.98 163\n", "\n", " accuracy 0.98 264\n", " macro avg 0.98 0.98 0.98 264\n", "weighted avg 0.98 0.98 0.98 264\n", "\n" ] } ], "source": [ "best_rf = grid_search.best_estimator_\n", "best_rf.fit(X_train, y_train)\n", "y_pred = best_rf.predict(X_test)\n", "\n", "# Evaluate performance\n", "from sklearn.metrics import accuracy_score, classification_report\n", "accuracy = accuracy_score(y_test, y_pred)\n", "print(\"Final Model Accuracy:\", accuracy)\n", "print(\"Classification Report:\\n\", classification_report(y_test, y_pred))" ] }, { "cell_type": "code", "execution_count": 1, "id": "e0752301-d13e-4ba9-82db-f1bf141efb2d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Selected Features: Index(['age', 'gender', 'glucose', 'kcm', 'troponin'], dtype='object')\n" ] } ], "source": [ "from sklearn.feature_selection import SelectKBest, chi2\n", "from sklearn.preprocessing import MinMaxScaler\n", "import pandas as pd\n", "\n", "# Load dataset\n", "data = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Separate features and target\n", "X = data.drop(\"class\", axis=1)\n", "y = (data[\"class\"] == \"positive\").astype(int) # Convert labels\n", "\n", "# Scale features for Chi-Square test\n", "scaler = MinMaxScaler()\n", "X_scaled = scaler.fit_transform(X)\n", "\n", "# Apply Chi-Square test for feature selection\n", "selector = SelectKBest(score_func=chi2, k=5) # Select top 5 features\n", "X_new = selector.fit_transform(X_scaled, y)\n", "\n", "# Show selected features\n", "selected_features = X.columns[selector.get_support()]\n", "print(\"Selected Features:\", selected_features)" ] }, { "cell_type": "code", "execution_count": 5, "id": "d62abff5-6e97-4c77-9ff1-7fdad005f88c", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.preprocessing import StandardScaler\n", "\n", "# Load dataset\n", "df = pd.read_csv(\"Heart Disease.csv\")\n", "\n", "# Define features (X) and target (y)\n", "X = df.drop(columns=[\"class\"]) # Assuming 'class' is the target variable\n", "y = df[\"class\"]\n", "\n", "# Convert categorical target variable to numerical (if needed)\n", "y = y.map({\"negative\": 0, \"positive\": 1})\n", "\n", "# Split into training and test sets\n", "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n", "\n", "# Standardize the features (important for models like Logistic Regression & SVM)\n", "scaler = StandardScaler()\n", "X_train = scaler.fit_transform(X_train)\n", "X_test = scaler.transform(X_test)" ] }, { "cell_type": "code", "execution_count": 7, "id": "3c24c325-deaa-4e33-a3bc-99949fce823d", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Selected Features: Index(['age', 'gender', 'glucose', 'kcm', 'troponin'], dtype='object')\n" ] } ], "source": [ "from sklearn.linear_model import LogisticRegression\n", "from sklearn.feature_selection import RFE\n", "\n", "# Initialize model\n", "model = LogisticRegression(max_iter=1000)\n", "\n", "# Apply Recursive Feature Elimination\n", "rfe = RFE(estimator=model, n_features_to_select=5) # Choose the number of features to keep\n", "X_train_rfe = rfe.fit_transform(X_train, y_train)\n", "X_test_rfe = rfe.transform(X_test)\n", "\n", "# Get selected feature names\n", "selected_features = X.columns[rfe.support_]\n", "print(\"Selected Features:\", selected_features)" ] }, { "cell_type": "code", "execution_count": 9, "id": "aa6a9715-93c8-430f-aade-e12836fb67d4", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model Accuracy after RFE: 0.80\n" ] } ], "source": [ "from sklearn.metrics import accuracy_score\n", "\n", "# Train the model on selected features\n", "model.fit(X_train_rfe, y_train)\n", "\n", "# Make predictions\n", " y_pred = model.predict(X_test_rfe)\n", " \n", " # Evaluate model performance\n", " accuracy = accuracy_score(y_test, y_pred)\n", " print(f\"Model Accuracy after RFE: {accuracy:.2f}\")" ] }, { "cell_type": "code", "execution_count": null, "id": "04a12b2c-5267-4cf7-a15e-63d657f345d0", "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 }