ml_rn.py

import numpy as np
import pandas as pd
import time
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.preprocessing import MinMaxScaler
from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score
from sklearn.neural_network import MLPRegressor

def load_and_describe_data(file_path):
    """
    Charge un fichier CSV et affiche les informations de base.
    """
    df = pd.read_csv(file_path)
    print(df.info())
    return df

def train_mlp(df):
    start_time = time.time()  # ⏳ Timer

    # 1️⃣ Séparation des features et de la cible
    X = df.drop(columns=["popularity", "id", "artists", "name", "release_date", "date_sortie", "duration_ms", "nom_artiste"])
    y = df["popularity"]

    # 2️⃣ Split train/test
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

    # 3️⃣ Normalisation des features
    scaler = MinMaxScaler()
    X_train_scaled = scaler.fit_transform(X_train)
    X_test_scaled = scaler.transform(X_test)

    # 4️⃣ Définition des hyperparamètres
    param_grid = {
        "hidden_layer_sizes": [(50,), (100,), (100, 50), (100, 100)],
        "activation": ["relu", "tanh"],
        "solver": ["adam", "sgd"],
        "learning_rate_init": [0.001, 0.01, 0.1],
        "max_iter": [500],
        "early_stopping": [True]  # Arrête si la validation ne s'améliore pas
    }

    # 5️⃣ Recherche des meilleurs hyperparamètres
    mlp = MLPRegressor(random_state=42)
    grid_search = GridSearchCV(mlp, param_grid, cv=3, scoring="r2", verbose=2)
    grid_search.fit(X_train_scaled, y_train)

    # 6️⃣ Affichage des meilleurs paramètres
    best_params = grid_search.best_params_
    print("\n✅ Meilleurs paramètres :", best_params)

    # 7️⃣ Prédiction avec le meilleur modèle
    best_mlp = grid_search.best_estimator_
    y_pred = best_mlp.predict(X_test_scaled)

    # 8️⃣ Évaluation du modèle
    mae = mean_absolute_error(y_test, y_pred)
    rmse = np.sqrt(mean_squared_error(y_test, y_pred))
    r2 = r2_score(y_test, y_pred)

    print(f"\n📊 MLPRegressor - MAE: {mae:.2f}, RMSE: {rmse:.2f}, R²: {r2:.3f}")

    # 9️⃣ Ajout des prédictions au DataFrame
    df.loc[X_test.index, "pred_mlp"] = y_pred

    # ⏳ Temps d'exécution
    elapsed_time = time.time() - start_time
    print(f"\n⏱️ Temps d'exécution : {elapsed_time:.2f} secondes")

    return df

# 📂 Chargement et entraînement
df = load_and_describe_data("data_sup_0popularity.csv")
df = train_mlp(df)