random_forest.py

#random_forest.py
from sklearn.ensemble import RandomForestRegressor
import pandas as pd
from sklearn.metrics import mean_absolute_error, r2_score


def rfr(X_train, y_train, X_test, y_test, features, stuff=''):
    #call the rfr model from sklearn
    model = RandomForestRegressor(
        n_estimators=200,
        max_depth=None,
        random_state=42,
        n_jobs=-1
    )

    #fit model
    model.fit(X_train, y_train)

    #predictions
    y_pred = model.predict(X_test)

    #Mean Absolute Error, difference between predicted and actual values
    mae = mean_absolute_error(y_test, y_pred)

    #R² score, predictions fit?
    r2 = r2_score(y_test, y_pred)

    print(f"Mean Absolute Error (MAE): {mae:.3f}")
    print(f"R² Score: {r2:.3f}")

    #save results
    r2_mae_results = pd.DataFrame({
        "Model": ["First Random Forest"],
        "Mean Absolute Error (MAE)": [mae],
        "R² Score/Mean R²": [r2]
    })
    #if stuff not null
    suffix = f"_{stuff}" if stuff else ""
    r2_mae_results.to_csv(f"data/r2_mae_results{suffix}.csv", index=False)

    #save results
    results = pd.DataFrame({
        "Actual": y_test,
        "Predicted": y_pred
    })
    results.to_csv(f"data/model_predictions{suffix}.csv", index=False)

    #which features affected predictions most
    importances = pd.DataFrame({
        "Feature": features,
        "Importance": model.feature_importances_
    }).sort_values(by="Importance", ascending=False)
    #save importantces to csv
    importances.to_csv(f"data/feature_importance{suffix}.csv", index=False)

    #print("\nFeature Importance:")
    #print(importances)