Refactor code for improved readability and consistency across multiple Jupyter notebooks

- Added missing commas in various print statements and function calls for better syntax.
- Reformatted code to enhance clarity, including breaking long lines and aligning parameters.
- Updated function signatures to use float type for sigma parameters instead of int for better precision.
- Cleaned up comments and documentation strings for clarity and consistency.
- Ensured consistent formatting in plotting functions and data handling.

M1/Statistical Learning/TP6_keras_intro.ipynb

@@ -92,6 +92,7 @@
    "outputs": [],
    "source": [
     "import numpy as np\n",
+    "\n",
     "import tensorflow as tf\n",
     "\n",
     "tf.keras.utils.set_random_seed(42)\n",
@@ -346,7 +347,7 @@
     "        tf.keras.layers.Dense(300, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(100, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")"
    ]
   },
@@ -691,7 +692,9 @@
    "outputs": [],
    "source": [
     "model.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")"
    ]
   },
@@ -1101,11 +1104,13 @@
     "        tf.keras.layers.Dense(300, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(100, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "\n",
     "model_10.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "\n",
     "model_10.fit(X_train01, y_train, epochs=10, validation_data=(X_val01, y_val))"
@@ -1270,7 +1275,8 @@
    ],
    "source": [
     "early_stopping_cb = tf.keras.callbacks.EarlyStopping(\n",
-    "    patience=5, restore_best_weights=True\n",
+    "    patience=5,\n",
+    "    restore_best_weights=True,\n",
     ")\n",
     "\n",
     "model = tf.keras.Sequential(\n",
@@ -1280,11 +1286,13 @@
     "        tf.keras.layers.Dense(300, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(100, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "\n",
     "model.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "\n",
     "history2 = model.fit(\n",
@@ -1598,10 +1606,12 @@
     "        tf.keras.layers.Input(shape=[28, 28]),\n",
     "        tf.keras.layers.Flatten(),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "reg_log.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "reg_log.fit(X_train01, y_train, epochs=90, validation_data=(X_val01, y_val))"
    ]
@@ -1709,10 +1719,12 @@
     "        tf.keras.layers.Dense(300, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(100, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "model_ter.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "model_ter.fit(X_train, y_train, epochs=30, validation_data=(X_val, y_val))"
    ]
@@ -1820,10 +1832,12 @@
     "        tf.keras.layers.Dense(300, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(100, activation=\"relu\", kernel_initializer=\"he_normal\"),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "model_5.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "\n",
     "X_train_far_too_small, X_val_far_too_small = X_train / 25500.0, X_val / 25500.0\n",
@@ -1938,16 +1952,22 @@
     "        tf.keras.layers.Input(shape=[28, 28]),\n",
     "        tf.keras.layers.Flatten(),\n",
     "        tf.keras.layers.Dense(\n",
-    "            300, activation=\"sigmoid\", kernel_initializer=\"he_normal\"\n",
+    "            300,\n",
+    "            activation=\"sigmoid\",\n",
+    "            kernel_initializer=\"he_normal\",\n",
     "        ),\n",
     "        tf.keras.layers.Dense(\n",
-    "            100, activation=\"sigmoid\", kernel_initializer=\"he_normal\"\n",
+    "            100,\n",
+    "            activation=\"sigmoid\",\n",
+    "            kernel_initializer=\"he_normal\",\n",
     "        ),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "model_sig_norm.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "model_sig_norm.fit(X_train01, y_train, epochs=30, validation_data=(X_val, y_val))"
    ]
@@ -2043,16 +2063,22 @@
     "        tf.keras.layers.Input(shape=[28, 28]),\n",
     "        tf.keras.layers.Flatten(),\n",
     "        tf.keras.layers.Dense(\n",
-    "            300, activation=\"sigmoid\", kernel_initializer=\"he_normal\"\n",
+    "            300,\n",
+    "            activation=\"sigmoid\",\n",
+    "            kernel_initializer=\"he_normal\",\n",
     "        ),\n",
     "        tf.keras.layers.Dense(\n",
-    "            100, activation=\"sigmoid\", kernel_initializer=\"he_normal\"\n",
+    "            100,\n",
+    "            activation=\"sigmoid\",\n",
+    "            kernel_initializer=\"he_normal\",\n",
     "        ),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "model_sig_un_norm.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "model_sig_un_norm.fit(X_train, y_train, epochs=30, validation_data=(X_val, y_val))"
    ]
@@ -2220,17 +2246,19 @@
     "        tf.keras.layers.Dense(300, activation=\"relu\"),\n",
     "        tf.keras.layers.Dense(100, activation=\"relu\"),\n",
     "        tf.keras.layers.Dense(10, activation=\"softmax\"),\n",
-    "    ]\n",
+    "    ],\n",
     ")\n",
     "model_high_variance.layers[1].set_weights(\n",
-    "    [200 * np.random.randn(28 * 28, 300) / 100, np.zeros(300)]\n",
+    "    [200 * np.random.randn(28 * 28, 300) / 100, np.zeros(300)],\n",
     ")\n",
     "model_high_variance.layers[2].set_weights(\n",
-    "    [200 * np.random.randn(300, 100) / 100, np.zeros(100)]\n",
+    "    [200 * np.random.randn(300, 100) / 100, np.zeros(100)],\n",
     ")\n",
     "\n",
     "model_high_variance.compile(\n",
-    "    loss=\"sparse_categorical_crossentropy\", optimizer=\"adam\", metrics=[\"accuracy\"]\n",
+    "    loss=\"sparse_categorical_crossentropy\",\n",
+    "    optimizer=\"adam\",\n",
+    "    metrics=[\"accuracy\"],\n",
     ")\n",
     "\n",
     "model_high_variance.fit(X_train01, y_train, epochs=60, validation_data=(X_val01, y_val))"