Leistungsnachweis-2/svm_modell.py

# Machine Learning Modell 1 / Leistungsnachweis II Data Science / 1. Semester
# Support Vector Machine zu binären Klassifikation

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split, cross_val_score, GridSearchCV
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.svm import SVC
from sklearn.metrics import classification_report, confusion_matrix, accuracy_score
from sklearn.impute import SimpleImputer
import matplotlib.pyplot as plt
import seaborn as sns
from datetime import datetime
import warnings
warnings.filterwarnings('ignore')

def load_and_preprocess_data():
    activities = pd.read_csv("Activities_rohdaten.csv", sep=None, engine="python")
    df = activities.copy()
    df.columns = df.columns.str.strip()

# Kategorisieren der Sportklassen
    def classify_activity(x):
        x = str(x).lower()
        if 'cardio' in x or 'gehen' in x or 'multisport' in x:
            return -1
        if 'kraft' in x:
            return 0
        if ('rad' in x or 'bike' in x or 'cycling' in x or 'velo' in x or 'schwimm' in x or 'laufen' in x or 'run' in x):
            return 1
        return -1

    df['sport_category'] = df['Aktivitätstyp'].apply(classify_activity)

    df = df[df['sport_category'] != -1]

# Numerische Daten welche miteinbezogen werden
    numeric_features = ['Distanz', 'Kalorien', 'Ø Herzfrequenz', 'Maximale Herzfrequenz', 'Aerober TE', 'Training Stress Score®']

    available_features = [col for col in numeric_features if col in df.columns]

    return df, available_features


def feature_engineering(df, features): 
    
    # Kopie erstellen
    df_processed = df.copy()
    
    # Numerische Features bereinigen
    for feature in features:
        if df_processed[feature].dtype == 'object':
            df_processed[feature] = df_processed[feature].astype(str).str.replace(',', '')
            
        # In numerische Werte umwandeln
        df_processed[feature] = pd.to_numeric(df_processed[feature], errors='coerce')
    
    # Zeitbasierte Features aus Datum extrahieren
    if 'Datum' in df_processed.columns:
        df_processed['Datum'] = pd.to_datetime(df_processed['Datum'], errors='coerce')
        df_processed['hour'] = df_processed['Datum'].dt.hour
        df_processed['day_of_week'] = df_processed['Datum'].dt.dayofweek
        df_processed['month'] = df_processed['Datum'].dt.month
        
        # Zeitbasierte Features hinzufügen
        time_features = ['hour', 'day_of_week', 'month']
        features.extend(time_features)
    
    # Fehlende Werte behandeln
    X = df_processed[features]
    y = df_processed['sport_category']
    
    # SimpleImputer für fehlende Werte
    imputer = SimpleImputer(strategy='median')
    X_imputed = imputer.fit_transform(X)
    
    return X_imputed, y, features, imputer

def train_svm_model(X, y):      # Trainiert das SVM-Modell mit Hyperparameter-Tuning
   
    # Daten in Trainings- und Testsets aufteilen
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42, stratify=y)
    
    # Features skalieren
    scaler = StandardScaler()
    X_train_scaled = scaler.fit_transform(X_train)
    X_test_scaled = scaler.transform(X_test)
    
    # SVM-Modell mit Hyperparameter-Tuning
    param_grid = {
        'C': [0.1, 1, 10, 100],
        'gamma': ['scale', 'auto', 0.001, 0.01, 0.1, 1],
        'kernel': ['rbf', 'linear']
    }
    
    svm = SVC(random_state=42)
    grid_search = GridSearchCV(svm, param_grid, cv=5, scoring='accuracy', n_jobs=-1)
    grid_search.fit(X_train_scaled, y_train)
    
    # Bestes Modell
    best_svm = grid_search.best_estimator_
    
    # Vorhersagen
    y_pred = best_svm.predict(X_test_scaled)
    
    # Kreuzvalidierung
    cv_scores = cross_val_score(best_svm, X_train_scaled, y_train, cv=5)
    
    return best_svm, X_test_scaled, y_test, y_pred, grid_search.best_params_, cv_scores, scaler

def evaluate_model(y_test, y_pred, cv_scores, best_params):         # Evaluiert das Modell und gibt Ergebnisse aus
    
    print("=== SVM-Modell für Sportarten-Klassifikation ===")
    print(f"Anzahl der Datensätze: {len(y_test) + len(y_pred)}")
    print(f"\nBeste Hyperparameter: {best_params}")
    
    print(f"\nKreuzvalidierung (CV-Scores): {cv_scores}")
    print(f"Mittlere CV-Genauigkeit: {cv_scores.mean():.4f} (+/- {cv_scores.std() * 2:.4f})")
    
    print(f"\nTest-Genauigkeit: {accuracy_score(y_test, y_pred):.4f}")
    
    print("\nKlassifikationsbericht:")
    print(classification_report(y_test, y_pred, target_names=['Kraftsport', 'Ausdauersport']))
    
    print("\nKonfusionsmatrix:")
    cm = confusion_matrix(y_test, y_pred)
    print(cm)
    
    # Visualisierung der Konfusionsmatrix
    plt.figure(figsize=(8, 6))
    sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', 
                xticklabels=['Kraftsport', 'Ausdauersport'],
                yticklabels=['Kraftsport', 'Ausdauersport'])
    plt.title('Konfusionsmatrix - SVM Sportarten-Klassifikation')
    plt.ylabel('Wahre Klasse')
    plt.xlabel('Vorhergesagte Klasse')
    plt.tight_layout()
    plt.savefig('confusion_matrix.png', dpi=300, bbox_inches='tight')
    plt.show()
    
    return accuracy_score(y_test, y_pred)

def feature_importance_analysis(model, X_test, feature_names):    # Analysiert die Wichtigkeit der Features (für lineare SVM)
    
    if model.kernel == 'linear':
        importance = np.abs(model.coef_[0])                       # Feature-Wichtigkeit für lineare SVM
        feature_importance = pd.DataFrame({'feature': feature_names, 'importance': importance}).sort_values('importance', ascending=False)
        
        print("\nFeature-Wichtigkeit (lineare SVM):")
        print(feature_importance)
      
        plt.figure(figsize=(10, 6))
        sns.barplot(data=feature_importance.head(10), x='importance', y='feature')
        plt.title('Top 10 Feature-Wichtigkeiten')
        plt.xlabel('Wichtigkeit')
        plt.tight_layout()
        plt.savefig('feature_importance.png', dpi=300, bbox_inches='tight')
        plt.show()
    else:
        print("\nFeature-Wichtigkeit nur für lineare SVM verfügbar")


def main():
    
    print("Starte SVM-Modell für Sportarten-Klassifikation...")
    
    try:
        # Daten laden und vorverarbeiten
        df, features = load_and_preprocess_data()
        print(f"Daten geladen: {df.shape[0]} Datensätze, {len(features)} Features")
        print(f"Verteilung der Sportarten: {df['sport_category'].value_counts().to_dict()}")
        
        # Feature Engineering
        X, y, feature_names, imputer = feature_engineering(df, features)
        print(f"Feature Engineering abgeschlossen: {X.shape[1]} Features")
        
        # SVM-Modell trainieren
        model, X_test, y_test, y_pred, best_params, cv_scores, scaler = train_svm_model(X, y)
        
        # Modell evaluieren
        accuracy = evaluate_model(y_test, y_pred, cv_scores, best_params)
        
        # Feature-Analyse
        feature_importance_analysis(model, X_test, feature_names)
        
        print(f"\nModell erfolgreich trainiert mit Genauigkeit: {accuracy:.4f}")
        print("\nGespicherte Dateien: confusion_matrix.png, feature_importance.png")
        
        return model, scaler, imputer, feature_names
        
    except Exception as e:
        print(f"Fehler bei der Ausführung: {str(e)}")
        return None, None, None, None

if __name__ == "__main__":
    model, scaler, imputer, feature_names = main()