mirror of
https://github.com/ArthurDanjou/ArtStudies.git
synced 2026-01-14 15:54:13 +01:00
Arthur DANJOU
8400c722a5
- Adjusted indentation and line breaks for better clarity in function definitions and import statements. - Standardized string quotes for consistency across the codebase. - Enhanced readability of DataFrame creation and manipulation by breaking long lines into multiple lines. - Cleaned up print statements and comments for improved understanding. - Ensured consistent use of whitespace around operators and after commas.
712 lines
351 KiB
Plaintext
712 lines
351 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Séance 3 - Compléments\n",
|
|
"\n",
|
|
"Dans cette séance nous travaillerons avec le dataset d'images [CIFAR10](https://keras.io/api/datasets/cifar10/) qui correspond à des petites images en couleurs. Notre objectif est de construire un réseau de neurones convolutionnel capable d'identifier chacun des dix types en exploitant quelque-unes des nouvelles méthodes décrites en cours.\n",
|
|
"\n",
|
|
"## Exploration des données\n",
|
|
"\n",
|
|
"Commençons par importer les données."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 134,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import numpy as np\n",
|
|
"import pandas as pd\n",
|
|
"\n",
|
|
"%matplotlib inline\n",
|
|
"import matplotlib.pyplot as plt\n",
|
|
"import seaborn as sns\n",
|
|
"\n",
|
|
"sns.set(style=\"whitegrid\")\n",
|
|
"\n",
|
|
"from tensorflow import keras\n",
|
|
"\n",
|
|
"(X_train, y_train), (X_valid, y_valid) = keras.datasets.cifar10.load_data()\n",
|
|
"\n",
|
|
"label_map = {\n",
|
|
" 0: \"airplane\",\n",
|
|
" 1: \"automobile\",\n",
|
|
" 2: \"bird\",\n",
|
|
" 3: \"cat\",\n",
|
|
" 4: \"deer\",\n",
|
|
" 5: \"dog\",\n",
|
|
" 6: \"frog\",\n",
|
|
" 7: \"horse\",\n",
|
|
" 8: \"ship\",\n",
|
|
" 9: \"truck \",\n",
|
|
"}"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Regardons la structure d'*y_train* avec son premier élément."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 135,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"[6]\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"print(y_train[0])"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Au lieu d'avoir un entier, nous avons un array. Pour pouvoir travailler, nous allons devoir modifier la structure de *y_train* et *y_valid*. Il faudrait passer d'un vecteur de taille $n$ à une matrice de type one-hot encoding de taille $(n, 10)$.\n",
|
|
"\n",
|
|
"**Consigne** : À l'aide de la fonction [`to_categorical`](https://keras.io/2.16/api/utils/python_utils/#tocategorical-function), modifier *y_train* et *y_valid*."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 136,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"[0. 0. 0. 0. 0. 0. 1. 0. 0. 0.]\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"y_train = keras.utils.to_categorical(y_train, num_classes=10)\n",
|
|
"y_valid = keras.utils.to_categorical(y_valid, num_classes=10)\n",
|
|
"\n",
|
|
"print(y_train[0])"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"**Consigne** : Afficher plusieurs images du dataset d'entraînement aléatoirement. On pourra utiliser la fonction [`imshow`](https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.imshow.html) et le dictionnaire."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 137,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1000x1000 with 25 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"plt.figure(figsize=(10, 10))\n",
|
|
"for i in range(25):\n",
|
|
" plt.subplot(5, 5, i + 1)\n",
|
|
" plt.xticks([])\n",
|
|
" plt.yticks([])\n",
|
|
" plt.grid(False)\n",
|
|
" plt.imshow(X_train[i])\n",
|
|
" plt.xlabel(label_map[np.argmax(y_train[i])])\n",
|
|
"plt.show()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"**Consigne** : Standardiser les données en utilisant la classe [`StandardScaler`](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html). On commencera par applatir les images en utilisant la méthode [`reshape`](https://numpy.org/doc/stable/reference/generated/numpy.reshape.html), puis on applique le pré-processing et on termine par reformer la matrice. Attention à bien respecter les dimensions d'origines de l'image."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 138,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"(50000, 32, 32, 3)\n",
|
|
"(10000, 32, 32, 3)\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"from sklearn.preprocessing import StandardScaler\n",
|
|
"\n",
|
|
"X_train_reshaped = X_train.reshape(-1, 32 * 32 * 3)\n",
|
|
"X_valid_reshaped = X_valid.reshape(-1, 32 * 32 * 3)\n",
|
|
"\n",
|
|
"scaler = StandardScaler()\n",
|
|
"X_train_scaled = scaler.fit_transform(X_train_reshaped).reshape(-1, 32, 32, 3)\n",
|
|
"X_valid_scaled = scaler.transform(X_valid_reshaped).reshape(-1, 32, 32, 3)\n",
|
|
"\n",
|
|
"X_train = X_train_scaled.astype(np.float32)\n",
|
|
"X_valid = X_valid_scaled.astype(np.float32)\n",
|
|
"\n",
|
|
"print(X_train.shape)\n",
|
|
"print(X_valid.shape)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Modélisation\n",
|
|
"\n",
|
|
"On souhaite visualiser les différentes courbes d'apprentissages obtenues par différent optimizer. Pour pouvoir le faire, nous allons devoir choisir les optimizers à comparer et lancer l'entraîner de plusieurs modèles. Commençons par définir une architecture avec une fonction de sorte à pouvoir simplement générer des modèles lors de la comparaisons entre les optimizers.\n",
|
|
"\n",
|
|
"**Consigne** : Définir une fonction `get_model` qui ne prend pas de paramètre et qui renvoie un modèle convolutionnel de moins de 200k paramètres en utilisant des couches de régularisations au choix."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 139,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_45\"</span>\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1mModel: \"sequential_45\"\u001b[0m\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
|
|
"┃<span style=\"font-weight: bold\"> Layer (type) </span>┃<span style=\"font-weight: bold\"> Output Shape </span>┃<span style=\"font-weight: bold\"> Param # </span>┃\n",
|
|
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
|
|
"│ conv2d_73 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ conv2d_74 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">9,248</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ max_pooling2d_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ conv2d_75 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">4,624</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ dropout_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ conv2d_76 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">2,320</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ flatten_45 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4096</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ dense_51 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">40,970</span> │\n",
|
|
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
|
|
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
|
|
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
|
|
"│ conv2d_73 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m896\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ conv2d_74 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m9,248\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ max_pooling2d_9 (\u001b[38;5;33mMaxPooling2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ conv2d_75 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m) │ \u001b[38;5;34m4,624\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ conv2d_76 (\u001b[38;5;33mConv2D\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m) │ \u001b[38;5;34m2,320\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ flatten_45 (\u001b[38;5;33mFlatten\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4096\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
|
|
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
|
|
"│ dense_51 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m) │ \u001b[38;5;34m40,970\u001b[0m │\n",
|
|
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">58,058</span> (226.79 KB)\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m58,058\u001b[0m (226.79 KB)\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">58,058</span> (226.79 KB)\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m58,058\u001b[0m (226.79 KB)\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
},
|
|
{
|
|
"data": {
|
|
"text/html": [
|
|
"<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
|
|
"</pre>\n"
|
|
],
|
|
"text/plain": [
|
|
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"def get_model() -> keras.Model:\n",
|
|
" model = keras.Sequential(\n",
|
|
" [\n",
|
|
" keras.layers.InputLayer(shape=(32, 32, 3)),\n",
|
|
" keras.layers.Conv2D(\n",
|
|
" filters=32, kernel_size=3, activation=\"relu\", padding=\"same\"\n",
|
|
" ),\n",
|
|
" keras.layers.Dropout(0.2),\n",
|
|
" keras.layers.Conv2D(\n",
|
|
" filters=32, kernel_size=3, activation=\"relu\", padding=\"same\"\n",
|
|
" ),\n",
|
|
" keras.layers.MaxPooling2D(pool_size=2),\n",
|
|
" keras.layers.Conv2D(\n",
|
|
" filters=16, kernel_size=3, activation=\"relu\", padding=\"same\"\n",
|
|
" ),\n",
|
|
" keras.layers.Dropout(0.2),\n",
|
|
" keras.layers.Conv2D(\n",
|
|
" filters=16, kernel_size=3, activation=\"relu\", padding=\"same\"\n",
|
|
" ),\n",
|
|
" keras.layers.Flatten(),\n",
|
|
" keras.layers.Dense(10, activation=\"softmax\"),\n",
|
|
" ]\n",
|
|
" )\n",
|
|
"\n",
|
|
" return model\n",
|
|
"\n",
|
|
"\n",
|
|
"model = get_model()\n",
|
|
"model.compile(optimizer=\"adam\", loss=\"categorical_crossentropy\", metrics=[\"accuracy\"])\n",
|
|
"model.summary()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Nous avons modifié la structure de *y_train* et *y_valid*, nous devons adapter la fonction de perte à optimiser en conséquence. Cette fois on considérera la fonction de perte [`CategoricalCrossentropy`](https://keras.io/api/losses/probabilistic_losses/#categoricalcrossentropy-class) au lieu de [`SparseCategoricalCrossentropy`](https://keras.io/api/losses/probabilistic_losses/#sparsecategoricalcrossentropy-class) que l'on utilisait jusqu'à présent.\n",
|
|
"\n",
|
|
"**Consigne** : Définir une fonction `compile_train` qui prend en paramètre:\n",
|
|
"* *optimizer_function* : l'instanciation de la classe de l'optimizer\n",
|
|
"* *learning_rate* : le learning rate associé à l'optimizer\n",
|
|
"* Et des [kwargs](https://book.pythontips.com/en/latest/args_and_kwargs.html)\n",
|
|
"\n",
|
|
"La fonction renvoie l'historique d'apprentissage du modèle définit par la fonction `get_model`. La fonction doit compiler le modèle avec l'optimizer définit en paramètre et l'entraîner avec les paramètres définit dans les kwargs."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 140,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def compile_train(\n",
|
|
" optimizer_function: str, learning_rate: float, **kwargs\n",
|
|
") -> keras.callbacks.History:\n",
|
|
" model = get_model()\n",
|
|
" optimizer = optimizer_function(learning_rate=learning_rate)\n",
|
|
" model.compile(\n",
|
|
" optimizer=optimizer,\n",
|
|
" loss=\"categorical_crossentropy\",\n",
|
|
" metrics=[\"accuracy\"],\n",
|
|
" )\n",
|
|
"\n",
|
|
" history = model.fit(\n",
|
|
" X_train,\n",
|
|
" y_train,\n",
|
|
" validation_data=(X_valid, y_valid),\n",
|
|
" **kwargs,\n",
|
|
" )\n",
|
|
"\n",
|
|
" return history"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"**Consigne** : Valider le bon fonctionnement de la fonction `compile_train` sur quelques époques."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 141,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 1/5\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 41ms/step - accuracy: 0.4744 - loss: 1.4691 - val_accuracy: 0.5455 - val_loss: 1.3027\n",
|
|
"Epoch 2/5\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m36s\u001b[0m 46ms/step - accuracy: 0.6046 - loss: 1.1210 - val_accuracy: 0.6342 - val_loss: 1.0331\n",
|
|
"Epoch 3/5\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.6577 - loss: 0.9842 - val_accuracy: 0.6728 - val_loss: 0.9386\n",
|
|
"Epoch 4/5\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 42ms/step - accuracy: 0.6874 - loss: 0.8941 - val_accuracy: 0.6752 - val_loss: 0.9314\n",
|
|
"Epoch 5/5\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 40ms/step - accuracy: 0.7079 - loss: 0.8316 - val_accuracy: 0.6953 - val_loss: 0.8628\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"epochs = 5\n",
|
|
"batch_size = 64\n",
|
|
"history_adam = compile_train(\n",
|
|
" keras.optimizers.Adam, learning_rate=0.001, epochs=epochs, batch_size=batch_size\n",
|
|
")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Pour s'affranchir un peu de l'aléatoire, nous proposons de lancer trois fois les différents schéma d'optimisation pour les comparer. La légende sera composée du nom de l'optimizer et la valeur du learning rate sélectionnée. La classe [`optimizer`](https://keras.io/api/optimizers/#optimizer-class) de Keras permet d'obtenir ces informations comme suit:"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 142,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Epoch 1/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m37s\u001b[0m 47ms/step - accuracy: 0.1829 - loss: 2.2094 - val_accuracy: 0.2651 - val_loss: 2.0630\n",
|
|
"Epoch 2/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m35s\u001b[0m 45ms/step - accuracy: 0.2754 - loss: 2.0188 - val_accuracy: 0.3196 - val_loss: 1.9536\n",
|
|
"Epoch 3/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m34s\u001b[0m 44ms/step - accuracy: 0.3111 - loss: 1.9346 - val_accuracy: 0.3469 - val_loss: 1.8800\n",
|
|
"Epoch 4/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.3397 - loss: 1.8597 - val_accuracy: 0.3724 - val_loss: 1.8069\n",
|
|
"Epoch 5/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m40s\u001b[0m 52ms/step - accuracy: 0.3697 - loss: 1.7821 - val_accuracy: 0.3944 - val_loss: 1.7373\n",
|
|
"Epoch 6/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 50ms/step - accuracy: 0.3966 - loss: 1.7081 - val_accuracy: 0.4239 - val_loss: 1.6589\n",
|
|
"Epoch 7/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m37s\u001b[0m 48ms/step - accuracy: 0.4165 - loss: 1.6502 - val_accuracy: 0.4155 - val_loss: 1.6464\n",
|
|
"Epoch 8/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m38s\u001b[0m 49ms/step - accuracy: 0.4320 - loss: 1.6071 - val_accuracy: 0.4418 - val_loss: 1.5936\n",
|
|
"Epoch 9/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m39s\u001b[0m 50ms/step - accuracy: 0.4436 - loss: 1.5670 - val_accuracy: 0.4469 - val_loss: 1.5567\n",
|
|
"Epoch 10/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.4554 - loss: 1.5379 - val_accuracy: 0.4548 - val_loss: 1.5364\n",
|
|
"Epoch 11/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 40ms/step - accuracy: 0.4625 - loss: 1.5074 - val_accuracy: 0.4708 - val_loss: 1.4966\n",
|
|
"Epoch 12/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 39ms/step - accuracy: 0.4734 - loss: 1.4824 - val_accuracy: 0.4719 - val_loss: 1.4689\n",
|
|
"Epoch 13/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.4810 - loss: 1.4587 - val_accuracy: 0.4737 - val_loss: 1.4651\n",
|
|
"Epoch 14/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m35s\u001b[0m 45ms/step - accuracy: 0.4863 - loss: 1.4405 - val_accuracy: 0.4917 - val_loss: 1.4227\n",
|
|
"Epoch 15/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 43ms/step - accuracy: 0.4927 - loss: 1.4178 - val_accuracy: 0.4971 - val_loss: 1.4096\n",
|
|
"Epoch 16/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.5056 - loss: 1.3945 - val_accuracy: 0.5002 - val_loss: 1.4000\n",
|
|
"Epoch 17/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.5094 - loss: 1.3763 - val_accuracy: 0.5175 - val_loss: 1.3561\n",
|
|
"Epoch 18/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 40ms/step - accuracy: 0.5164 - loss: 1.3565 - val_accuracy: 0.5179 - val_loss: 1.3505\n",
|
|
"Epoch 19/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 35ms/step - accuracy: 0.5240 - loss: 1.3405 - val_accuracy: 0.5316 - val_loss: 1.3356\n",
|
|
"Epoch 20/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m30s\u001b[0m 38ms/step - accuracy: 0.5286 - loss: 1.3256 - val_accuracy: 0.5422 - val_loss: 1.3064\n",
|
|
"Epoch 1/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 41ms/step - accuracy: 0.4644 - loss: 1.4877 - val_accuracy: 0.5663 - val_loss: 1.2312\n",
|
|
"Epoch 2/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.6058 - loss: 1.1150 - val_accuracy: 0.6273 - val_loss: 1.0454\n",
|
|
"Epoch 3/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m34s\u001b[0m 43ms/step - accuracy: 0.6596 - loss: 0.9730 - val_accuracy: 0.6772 - val_loss: 0.9169\n",
|
|
"Epoch 4/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m35s\u001b[0m 45ms/step - accuracy: 0.6894 - loss: 0.8917 - val_accuracy: 0.6930 - val_loss: 0.8928\n",
|
|
"Epoch 5/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7107 - loss: 0.8324 - val_accuracy: 0.7017 - val_loss: 0.8571\n",
|
|
"Epoch 6/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 38ms/step - accuracy: 0.7247 - loss: 0.7947 - val_accuracy: 0.7141 - val_loss: 0.8377\n",
|
|
"Epoch 7/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m27s\u001b[0m 34ms/step - accuracy: 0.7338 - loss: 0.7605 - val_accuracy: 0.7150 - val_loss: 0.8346\n",
|
|
"Epoch 8/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m27s\u001b[0m 34ms/step - accuracy: 0.7443 - loss: 0.7337 - val_accuracy: 0.7139 - val_loss: 0.8275\n",
|
|
"Epoch 9/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m27s\u001b[0m 35ms/step - accuracy: 0.7521 - loss: 0.7063 - val_accuracy: 0.7054 - val_loss: 0.8556\n",
|
|
"Epoch 10/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.7584 - loss: 0.6912 - val_accuracy: 0.7271 - val_loss: 0.7955\n",
|
|
"Epoch 11/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.7619 - loss: 0.6771 - val_accuracy: 0.7191 - val_loss: 0.8228\n",
|
|
"Epoch 12/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 35ms/step - accuracy: 0.7693 - loss: 0.6571 - val_accuracy: 0.7197 - val_loss: 0.8309\n",
|
|
"Epoch 13/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 35ms/step - accuracy: 0.7726 - loss: 0.6451 - val_accuracy: 0.7264 - val_loss: 0.8150\n",
|
|
"Epoch 14/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 39ms/step - accuracy: 0.7767 - loss: 0.6338 - val_accuracy: 0.7205 - val_loss: 0.8179\n",
|
|
"Epoch 15/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7801 - loss: 0.6237 - val_accuracy: 0.7264 - val_loss: 0.8154\n",
|
|
"Epoch 16/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.7849 - loss: 0.6153 - val_accuracy: 0.7261 - val_loss: 0.8193\n",
|
|
"Epoch 17/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 35ms/step - accuracy: 0.7835 - loss: 0.6121 - val_accuracy: 0.7224 - val_loss: 0.8252\n",
|
|
"Epoch 18/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 36ms/step - accuracy: 0.7898 - loss: 0.5957 - val_accuracy: 0.7174 - val_loss: 0.8437\n",
|
|
"Epoch 19/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.7919 - loss: 0.5912 - val_accuracy: 0.7205 - val_loss: 0.8248\n",
|
|
"Epoch 20/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7963 - loss: 0.5821 - val_accuracy: 0.7289 - val_loss: 0.8188\n",
|
|
"Epoch 1/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m36s\u001b[0m 43ms/step - accuracy: 0.4750 - loss: 1.4634 - val_accuracy: 0.5656 - val_loss: 1.2143\n",
|
|
"Epoch 2/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 39ms/step - accuracy: 0.6092 - loss: 1.1060 - val_accuracy: 0.6366 - val_loss: 1.0502\n",
|
|
"Epoch 3/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 39ms/step - accuracy: 0.6613 - loss: 0.9752 - val_accuracy: 0.6667 - val_loss: 0.9622\n",
|
|
"Epoch 4/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.6907 - loss: 0.8947 - val_accuracy: 0.6877 - val_loss: 0.9018\n",
|
|
"Epoch 5/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m35s\u001b[0m 44ms/step - accuracy: 0.7055 - loss: 0.8443 - val_accuracy: 0.6965 - val_loss: 0.8762\n",
|
|
"Epoch 6/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7221 - loss: 0.8018 - val_accuracy: 0.7020 - val_loss: 0.8675\n",
|
|
"Epoch 7/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7302 - loss: 0.7722 - val_accuracy: 0.7136 - val_loss: 0.8310\n",
|
|
"Epoch 8/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 37ms/step - accuracy: 0.7408 - loss: 0.7435 - val_accuracy: 0.7117 - val_loss: 0.8283\n",
|
|
"Epoch 9/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7504 - loss: 0.7161 - val_accuracy: 0.7170 - val_loss: 0.8168\n",
|
|
"Epoch 10/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m32s\u001b[0m 41ms/step - accuracy: 0.7559 - loss: 0.7003 - val_accuracy: 0.7190 - val_loss: 0.8165\n",
|
|
"Epoch 11/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m26s\u001b[0m 33ms/step - accuracy: 0.7619 - loss: 0.6823 - val_accuracy: 0.7130 - val_loss: 0.8485\n",
|
|
"Epoch 12/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m31s\u001b[0m 40ms/step - accuracy: 0.7681 - loss: 0.6654 - val_accuracy: 0.7204 - val_loss: 0.8313\n",
|
|
"Epoch 13/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 36ms/step - accuracy: 0.7689 - loss: 0.6538 - val_accuracy: 0.7180 - val_loss: 0.8248\n",
|
|
"Epoch 14/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m33s\u001b[0m 42ms/step - accuracy: 0.7717 - loss: 0.6444 - val_accuracy: 0.7166 - val_loss: 0.8376\n",
|
|
"Epoch 15/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m29s\u001b[0m 38ms/step - accuracy: 0.7801 - loss: 0.6282 - val_accuracy: 0.7204 - val_loss: 0.8239\n",
|
|
"Epoch 16/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m30s\u001b[0m 39ms/step - accuracy: 0.7811 - loss: 0.6227 - val_accuracy: 0.7222 - val_loss: 0.8318\n",
|
|
"Epoch 17/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 36ms/step - accuracy: 0.7855 - loss: 0.6114 - val_accuracy: 0.7244 - val_loss: 0.8211\n",
|
|
"Epoch 18/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m25s\u001b[0m 32ms/step - accuracy: 0.7857 - loss: 0.6033 - val_accuracy: 0.7157 - val_loss: 0.8554\n",
|
|
"Epoch 19/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m28s\u001b[0m 36ms/step - accuracy: 0.7867 - loss: 0.5988 - val_accuracy: 0.7271 - val_loss: 0.8364\n",
|
|
"Epoch 20/20\n",
|
|
"\u001b[1m782/782\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m27s\u001b[0m 34ms/step - accuracy: 0.7910 - loss: 0.5856 - val_accuracy: 0.7243 - val_loss: 0.8319\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"learning_rate = 0.001\n",
|
|
"epochs = 20\n",
|
|
"batch_size = 64\n",
|
|
"optimizers = [\n",
|
|
" keras.optimizers.SGD,\n",
|
|
" keras.optimizers.AdamW,\n",
|
|
" keras.optimizers.Adam,\n",
|
|
"]\n",
|
|
"\n",
|
|
"histories = []\n",
|
|
"for optimizer in optimizers:\n",
|
|
" history = compile_train(\n",
|
|
" optimizer, learning_rate=learning_rate, epochs=epochs, batch_size=batch_size\n",
|
|
" )\n",
|
|
" name = optimizer.__name__\n",
|
|
" label = f\"{name} (lr={learning_rate:.06})\"\n",
|
|
" histories.append({\"label\": label, \"history\": history})"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 143,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"data": {
|
|
"image/png": "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",
|
|
"text/plain": [
|
|
"<Figure size 1000x1000 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"plt.figure(figsize=(10, 10))\n",
|
|
"colors = sns.color_palette(\"husl\", len(histories))\n",
|
|
"for i, record in enumerate(histories):\n",
|
|
" label = record[\"label\"]\n",
|
|
" history = record[\"history\"]\n",
|
|
"\n",
|
|
" loss = history.history[\"loss\"]\n",
|
|
" val_loss = history.history[\"val_loss\"]\n",
|
|
"\n",
|
|
" mean_train = np.mean(loss)\n",
|
|
" std_train = np.std(loss)\n",
|
|
" mean_val = np.mean(val_loss)\n",
|
|
" std_val = np.std(val_loss)\n",
|
|
"\n",
|
|
" plt.plot(history.history[\"loss\"], label=label, color=colors[i], alpha=0.4)\n",
|
|
" plt.fill_between(\n",
|
|
" range(len(loss)),\n",
|
|
" np.array(loss) - np.array(std_train),\n",
|
|
" np.array(loss) + np.array(std_train),\n",
|
|
" color=colors[i],\n",
|
|
" alpha=0.1,\n",
|
|
" )\n",
|
|
" plt.plot(history.history[\"val_loss\"], linestyle=\"--\", color=colors[i], linewidth=3)\n",
|
|
" plt.fill_between(\n",
|
|
" range(len(val_loss)),\n",
|
|
" np.array(val_loss) - np.array(std_val),\n",
|
|
" np.array(val_loss) + np.array(std_val),\n",
|
|
" color=colors[i],\n",
|
|
" alpha=0.1,\n",
|
|
" )\n",
|
|
"plt.title(\"Validation Loss by Optimizer\")\n",
|
|
"plt.xlabel(\"Epochs\")\n",
|
|
"plt.ylabel(\"Validation Loss\")\n",
|
|
"plt.legend()\n",
|
|
"plt.show()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"**Consigne** : Écrire une boucle d'entraînement qui va stocker dans une liste les courbes d'apprentissage. Chaque élément de la liste correspondra à un dictionnaire avec pour clé:\n",
|
|
"* *type*: le nom de l'optimizer\n",
|
|
"* *history*: l'historique d'apprentissage"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"Il faut maintenant visualiser les résultats. Commençons par préparer les données.\n",
|
|
"\n",
|
|
"**Consigne** : Définir une fonction `agregate_result` qui prend en paramètre:\n",
|
|
"* *results*: le dictionnaire de résultat, au format décrit précédemment\n",
|
|
"* *network_type*: chaîne de caractère identifiant le type de réseau\n",
|
|
"* *metric_name*: le nom de la métrique d'intérêt\n",
|
|
"\n",
|
|
"La fonction renverra deux matrices de tailles (nombre de comparaisons, nombre d'époque) : une pour le dataset d'entraînement et une pour le dataset de validation. On concatène donc les différentes courbes d'apprentissage."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"**Consigne** : Visualiser les courbes d'apprentissage en faisant apparaître des intervals de confiance. On prendra exemple sur la fonction `show_results` du TP précédent. Commenter."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Pour continuer\n",
|
|
"\n",
|
|
"Choisir une ou plusieurs pistes de recherche parmi les suivantes. Il est possible de choisir une autre direction, mais elle doit être validé auparavant.\n",
|
|
"\n",
|
|
"1. Nous avons utilisé un learning rate fixe et dans le cours nous avons parlé d'échéancier. Comparer les deux approches, puis se poser la question de l'importance (ou non) d'un phase de warmup.\n",
|
|
"2. Le [`Dropout`](https://keras.io/api/layers/regularization_layers/dropout/) permet de régulariser un réseau de neurones. Comparer un réseau avec et sans dropout, puis se poser la question de l'importance de la magnitude et du placement d'une couche dropout."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "studies",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.13.3"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|