editing tp3 encoder main file by adding commentary

exos/tp3_autoencoder/main.py

@@ -4,24 +4,32 @@ from keras.models import Model
 from keras.layers import Input, Dense
 from keras.optimizers import Adam
 
-# Charger et préparer les données
+
 data = np.loadtxt('datas/semeion.data')
 
-# Séparer les features et les labels
 X = data[:, :-10]  # Ignorer les 10 dernières colonnes
-y = data[:, -10:]
-
-# Diviser les données en ensembles d'entraînement et de test
-X_train, X_test = train_test_split(X, test_size=0.2, random_state=42)
+y = data[:, -10:] # Les 10 dernières colonnes sont les étiquettes
+X_train, X_test = train_test_split(X, test_size=0.2, random_state=42) # Diviser les données en ensembles d'entraînement et de test
 
-# Fonction pour ajouter du bruit aux données
 def add_noise(data, noise_factor=0.5):
+    """
+    Add random noise to the data. The noise factor controls the amount of noise added. The noisy data is clipped to ensure it falls within the valid range [0, 1].
+    :param data: numpy.ndarray: The input data.
+    :param noise_factor: float: The noise factor.
+    :return: numpy.ndarray: The noisy data.
+    """
     noisy_data = data + noise_factor * np.random.normal(loc=0.0, scale=1.0, size=data.shape)
     noisy_data = np.clip(noisy_data, 0., 1.)
     return noisy_data
 
 # Auto-encodeur avec une couche cachée
 def build_single_layer_autoencoder(input_dim, hidden_dim):
+    """
+    Build an autoencoder with a single hidden layer.
+    :param input_dim:
+    :param hidden_dim:
+    :return: keras.models.Model: The autoencoder model.
+    """
     input_layer = Input(shape=(input_dim,))
     hidden_layer = Dense(hidden_dim, activation='relu')(input_layer)
     output_layer = Dense(input_dim, activation='sigmoid')(hidden_layer)
@@ -31,6 +39,12 @@ def build_single_layer_autoencoder(input_dim, hidden_dim):
 
 # Auto-encodeur avec trois couches cachées
 def build_three_layer_autoencoder(input_dim, hidden_dims):
+    """
+    Build an autoencoder with three hidden layers.
+    :param input_dim:
+    :param hidden_dims:
+    :return: keras.models.Model: The autoencoder model.
+    """
     input_layer = Input(shape=(input_dim,))
     hidden_layer1 = Dense(hidden_dims[0], activation='relu')(input_layer)
     hidden_layer2 = Dense(hidden_dims[1], activation='relu')(hidden_layer1)
@@ -42,6 +56,14 @@ def build_three_layer_autoencoder(input_dim, hidden_dims):
 
 # Entraîner et évaluer un auto-encodeur
 def train_and_evaluate_autoencoder(autoencoder, X_train, X_test, epochs=50, batch_size=256):
+    """
+    Train and evaluate an autoencoder model on the given data.
+    :param autoencoder: the autoencoder model to train and evaluate
+    :param X_train: training data
+    :param X_test: test data
+    :param epochs: number of epochs to train the model
+    :param batch_size: batch size for training
+    """
     autoencoder.fit(X_train, X_train, epochs=epochs, batch_size=batch_size, shuffle=True, validation_data=(X_test, X_test))
     reconstructed = autoencoder.predict(X_test)
     mse = np.mean(np.power(X_test - reconstructed, 2), axis=1)
@@ -49,6 +71,16 @@ def train_and_evaluate_autoencoder(autoencoder, X_train, X_test, epochs=50, batc
 
 # Entraîner et évaluer un auto-encodeur avec des données bruitées
 def train_and_evaluate_noisy_autoencoder(autoencoder, X_train, X_test, noise_factor=0.5, epochs=50, batch_size=256):
+    """
+    Train and evaluate an autoencoder model on noisy data.
+    :param autoencoder: the autoencoder model to train and evaluate
+    :param X_train: training data without noise (clean data)
+    :param X_test: test data without noise (clean data)
+    :param noise_factor: the factor controlling the amount of noise to add to the data
+    :param epochs: number of epochs to train the model
+    :param batch_size: batch size for training
+    :return:
+    """
     X_train_noisy = add_noise(X_train, noise_factor)
     X_test_noisy = add_noise(X_test, noise_factor)
     autoencoder.fit(X_train_noisy, X_train, epochs=epochs, batch_size=batch_size, shuffle=True, validation_data=(X_test_noisy, X_test))
@@ -56,7 +88,7 @@ def train_and_evaluate_noisy_autoencoder(autoencoder, X_train, X_test, noise_fac
     mse = np.mean(np.power(X_test - reconstructed, 2), axis=1)
     print("Mean Squared Error with noise: ", np.mean(mse))
 
-# Paramètres
+# Paramètres pour les auto-encodeurs à une et trois couches cachées
 input_dim = X_train.shape[1]
 
 # Entraîner et évaluer l'auto-encodeur avec une couche cachée