mirror of
https://github.com/ArthurDanjou/breast-cancer-detection.git
synced 2026-01-14 15:54:14 +01:00
535 lines
255 KiB
Plaintext
535 lines
255 KiB
Plaintext
{
|
||
"cells": [
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 1,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"import numpy as np \n",
|
||
"import pandas as pd\n",
|
||
"import tensorflow as tf\n",
|
||
"import matplotlib.pyplot as plt"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 2,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
" Age BMI Glucose Insulin HOMA Leptin Adiponectin Resistin \\\n",
|
||
"0 48 23.500000 70 2.707 0.467409 8.8071 9.702400 7.99585 \n",
|
||
"1 83 20.690495 92 3.115 0.706897 8.8438 5.429285 4.06405 \n",
|
||
"2 82 23.124670 91 4.498 1.009651 17.9393 22.432040 9.27715 \n",
|
||
"3 68 21.367521 77 3.226 0.612725 9.8827 7.169560 12.76600 \n",
|
||
"4 86 21.111111 92 3.549 0.805386 6.6994 4.819240 10.57635 \n",
|
||
"\n",
|
||
" MCP.1 Classification \n",
|
||
"0 417.114 1 \n",
|
||
"1 468.786 1 \n",
|
||
"2 554.697 1 \n",
|
||
"3 928.220 1 \n",
|
||
"4 773.920 1 \n",
|
||
" Age BMI Glucose Insulin HOMA Leptin \\\n",
|
||
"count 116.000000 116.000000 116.000000 116.000000 116.000000 116.000000 \n",
|
||
"mean 57.301724 27.582111 97.793103 10.012086 2.694988 26.615080 \n",
|
||
"std 16.112766 5.020136 22.525162 10.067768 3.642043 19.183294 \n",
|
||
"min 24.000000 18.370000 60.000000 2.432000 0.467409 4.311000 \n",
|
||
"25% 45.000000 22.973205 85.750000 4.359250 0.917966 12.313675 \n",
|
||
"50% 56.000000 27.662416 92.000000 5.924500 1.380939 20.271000 \n",
|
||
"75% 71.000000 31.241442 102.000000 11.189250 2.857787 37.378300 \n",
|
||
"max 89.000000 38.578759 201.000000 58.460000 25.050342 90.280000 \n",
|
||
"\n",
|
||
" Adiponectin Resistin MCP.1 Classification \n",
|
||
"count 116.000000 116.000000 116.000000 116.000000 \n",
|
||
"mean 10.180874 14.725966 534.647000 1.551724 \n",
|
||
"std 6.843341 12.390646 345.912663 0.499475 \n",
|
||
"min 1.656020 3.210000 45.843000 1.000000 \n",
|
||
"25% 5.474283 6.881763 269.978250 1.000000 \n",
|
||
"50% 8.352692 10.827740 471.322500 2.000000 \n",
|
||
"75% 11.815970 17.755207 700.085000 2.000000 \n",
|
||
"max 38.040000 82.100000 1698.440000 2.000000 \n",
|
||
"<class 'pandas.core.frame.DataFrame'>\n",
|
||
"RangeIndex: 116 entries, 0 to 115\n",
|
||
"Data columns (total 10 columns):\n",
|
||
" # Column Non-Null Count Dtype \n",
|
||
"--- ------ -------------- ----- \n",
|
||
" 0 Age 116 non-null int64 \n",
|
||
" 1 BMI 116 non-null float64\n",
|
||
" 2 Glucose 116 non-null int64 \n",
|
||
" 3 Insulin 116 non-null float64\n",
|
||
" 4 HOMA 116 non-null float64\n",
|
||
" 5 Leptin 116 non-null float64\n",
|
||
" 6 Adiponectin 116 non-null float64\n",
|
||
" 7 Resistin 116 non-null float64\n",
|
||
" 8 MCP.1 116 non-null float64\n",
|
||
" 9 Classification 116 non-null int64 \n",
|
||
"dtypes: float64(7), int64(3)\n",
|
||
"memory usage: 9.2 KB\n",
|
||
"None\n",
|
||
"Age 0\n",
|
||
"BMI 0\n",
|
||
"Glucose 0\n",
|
||
"Insulin 0\n",
|
||
"HOMA 0\n",
|
||
"Leptin 0\n",
|
||
"Adiponectin 0\n",
|
||
"Resistin 0\n",
|
||
"MCP.1 0\n",
|
||
"Classification 0\n",
|
||
"dtype: int64\n"
|
||
]
|
||
},
|
||
{
|
||
"data": {
|
||
"text/plain": [
|
||
"array([[<Axes: title={'center': 'Age'}>, <Axes: title={'center': 'BMI'}>,\n",
|
||
" <Axes: title={'center': 'Glucose'}>],\n",
|
||
" [<Axes: title={'center': 'Insulin'}>,\n",
|
||
" <Axes: title={'center': 'HOMA'}>,\n",
|
||
" <Axes: title={'center': 'Leptin'}>],\n",
|
||
" [<Axes: title={'center': 'Adiponectin'}>,\n",
|
||
" <Axes: title={'center': 'Resistin'}>,\n",
|
||
" <Axes: title={'center': 'MCP.1'}>],\n",
|
||
" [<Axes: title={'center': 'Classification'}>, <Axes: >, <Axes: >]],\n",
|
||
" dtype=object)"
|
||
]
|
||
},
|
||
"execution_count": 2,
|
||
"metadata": {},
|
||
"output_type": "execute_result"
|
||
},
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 1000x1000 with 12 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"breast_cancer = pd.read_csv(\"data/breast_cancer.csv\")\n",
|
||
"print(breast_cancer.head())\n",
|
||
"\n",
|
||
"print(breast_cancer.describe())\n",
|
||
"\n",
|
||
"print(breast_cancer.info())\n",
|
||
"\n",
|
||
"print(breast_cancer.isnull().sum())\n",
|
||
"\n",
|
||
"breast_cancer.hist(figsize=(10, 10))"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 3,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Classification\n",
|
||
"1 64\n",
|
||
"0 52\n",
|
||
"Name: count, dtype: int64\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"X = breast_cancer.drop(columns=[\"Classification\"])\n",
|
||
"y = breast_cancer[\"Classification\"].replace({1: 0, 2: 1})\n",
|
||
"\n",
|
||
"print(y.value_counts())"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 4,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"Number of samples: 116\n",
|
||
"Number of features: 9\n",
|
||
"Number of classes: 2\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"print(\"Number of samples:\", X.shape[0])\n",
|
||
"print(\"Number of features:\", X.shape[1])\n",
|
||
"print(\"Number of classes:\", len(np.unique(y)))"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 15,
|
||
"metadata": {},
|
||
"outputs": [],
|
||
"source": [
|
||
"def build_model():\n",
|
||
" model = tf.keras.models.Sequential([\n",
|
||
" tf.keras.layers.Dense(16, activation='relu', input_shape=(X.shape[1],), kernel_regularizer=tf.keras.regularizers.l2(0.01)),\n",
|
||
" tf.keras.layers.Dense(8, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(0.01)),\n",
|
||
" tf.keras.layers.Dense(1, activation='sigmoid')\n",
|
||
" ])\n",
|
||
" model.compile(\n",
|
||
" optimizer='adam',\n",
|
||
" loss='binary_crossentropy',\n",
|
||
" metrics=['accuracy']\n",
|
||
" )\n",
|
||
" return model"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 18,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/Users/arthurdanjou/Workspace/studies/.venv/lib/python3.12/site-packages/keras/src/layers/core/dense.py:93: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
|
||
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n",
|
||
"Fold 1 - F1-score : 0.7333\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/Users/arthurdanjou/Workspace/studies/.venv/lib/python3.12/site-packages/keras/src/layers/core/dense.py:93: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
|
||
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
|
||
"Fold 2 - F1-score : 0.6667\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/Users/arthurdanjou/Workspace/studies/.venv/lib/python3.12/site-packages/keras/src/layers/core/dense.py:93: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
|
||
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n",
|
||
"Fold 3 - F1-score : 0.7222\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/Users/arthurdanjou/Workspace/studies/.venv/lib/python3.12/site-packages/keras/src/layers/core/dense.py:93: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
|
||
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
|
||
"Fold 4 - F1-score : 0.9231\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/Users/arthurdanjou/Workspace/studies/.venv/lib/python3.12/site-packages/keras/src/layers/core/dense.py:93: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
|
||
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 50ms/step\n",
|
||
"Fold 5 - F1-score : 0.7586\n",
|
||
"\n",
|
||
"F1-score moyen sur 5 folds : 0.7608\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"from sklearn.metrics import f1_score\n",
|
||
"from sklearn.model_selection import StratifiedKFold\n",
|
||
"from keras.callbacks import EarlyStopping\n",
|
||
"from sklearn.preprocessing import StandardScaler\n",
|
||
"\n",
|
||
"\n",
|
||
"skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=42)\n",
|
||
"f1_scores = []\n",
|
||
"histories = []\n",
|
||
"\n",
|
||
"early_stopping = EarlyStopping(\n",
|
||
" monitor='val_loss',\n",
|
||
" patience=10,\n",
|
||
" restore_best_weights=True,\n",
|
||
" verbose=1\n",
|
||
")\n",
|
||
"\n",
|
||
"for fold, (train_idx, val_idx) in enumerate(skf.split(X, y), 1):\n",
|
||
" X_train, X_val = X.iloc[train_idx], X.iloc[val_idx]\n",
|
||
" y_train, y_val = y.iloc[train_idx], y.iloc[val_idx]\n",
|
||
"\n",
|
||
" # Standardisation\n",
|
||
" scaler = StandardScaler()\n",
|
||
" X_train_scaled = scaler.fit_transform(X_train)\n",
|
||
" X_val_scaled = scaler.transform(X_val)\n",
|
||
" \n",
|
||
" model = build_model()\n",
|
||
"\n",
|
||
" model.compile(\n",
|
||
" optimizer='adam',\n",
|
||
" loss='binary_crossentropy',\n",
|
||
" metrics=[\"f1_score\"]\n",
|
||
" )\n",
|
||
"\n",
|
||
" # EarlyStopping\n",
|
||
" callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n",
|
||
"\n",
|
||
" # Entraînement\n",
|
||
" history = model.fit(\n",
|
||
" X_train_scaled, y_train,\n",
|
||
" epochs=50,\n",
|
||
" batch_size=8,\n",
|
||
" validation_data=(X_val_scaled, y_val),\n",
|
||
" callbacks=[callback],\n",
|
||
" verbose=0,\n",
|
||
" class_weight={0: 1.0, 1: 2.0}\n",
|
||
" )\n",
|
||
" \n",
|
||
" histories.append(history.history)\n",
|
||
"\n",
|
||
" # Prédiction & F1\n",
|
||
" y_pred_val = (model.predict(X_val_scaled) > 0.5).astype(int)\n",
|
||
" score = f1_score(y_val, y_pred_val)\n",
|
||
" f1_scores.append(score)\n",
|
||
" print(f\"Fold {fold} - F1-score : {score:.4f}\")\n",
|
||
"\n",
|
||
"print(f\"\\nF1-score moyen sur 5 folds : {np.mean(f1_scores):.4f}\")"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 19,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 1200x1200 with 5 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"fig, axes = plt.subplots(3, 2, figsize=(12, 12), sharey=True) # 3 rows, 2 columns\n",
|
||
"axes = axes.flatten() # Flatten to easily iterate\n",
|
||
"\n",
|
||
"for i, (hist, ax) in enumerate(zip(histories, axes)):\n",
|
||
" ax.plot(hist['loss'], label='Train loss', alpha=0.6)\n",
|
||
" ax.plot(hist['val_loss'], label='Val loss', linestyle='--', alpha=0.6)\n",
|
||
" ax.set_title(f\"Fold {i+1}\")\n",
|
||
" ax.set_xlabel(\"Epochs\")\n",
|
||
" if i % 2 == 0:\n",
|
||
" ax.set_ylabel(\"Binary Crossentropy\")\n",
|
||
" ax.legend(fontsize=8)\n",
|
||
" ax.grid(True)\n",
|
||
"\n",
|
||
"# Hide any unused subplots if histories < 6\n",
|
||
"for j in range(len(histories), len(axes)):\n",
|
||
" fig.delaxes(axes[j])\n",
|
||
"\n",
|
||
"plt.suptitle(\"Évolution de la loss sur chaque fold\")\n",
|
||
"plt.tight_layout(rect=[0, 0, 1, 0.96])\n",
|
||
"plt.show()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {},
|
||
"source": [
|
||
"Dans le cadre de la classification binaire à partir du dataset Breast Cancer Coimbra, le respect de la distribution des classes lors de la séparation des données est une condition essentielle à la validité des résultats expérimentaux.\n",
|
||
"\n",
|
||
"1. Utilisation de stratify=y dans train_test_split\n",
|
||
"Lors de la séparation du jeu de données en un ensemble d'entraînement et un ensemble de test, nous avons recours à la fonction train_test_split de la bibliothèque scikit-learn. Afin de garantir que les proportions des classes cibles soient conservées dans les deux sous-ensembles, l’argument stratify=y est utilisé.\n",
|
||
"\n",
|
||
"Cette précaution est particulièrement importante dans le cas de jeux de données déséquilibrés, comme c’est le cas ici, où les deux classes de la variable cible (\"Classification\") ne sont pas également représentées. Un échantillonnage aléatoire simple pourrait introduire un déséquilibre important entre les classes dans le jeu de test, rendant les métriques de performance peu fiables et favorisant potentiellement une classe au détriment de l’autre. L’option stratify=y assure donc une représentativité statistique des classes dans chacun des sous-échantillons.\n",
|
||
"\n",
|
||
"2. Recours à StratifiedKFold pour la validation croisée\n",
|
||
"De manière analogue, lors de l'évaluation du modèle par validation croisée, nous avons choisi l’utilisation de la méthode StratifiedKFold. Contrairement à la validation croisée standard (KFold), cette méthode garantit que la proportion des classes est maintenue dans chacun des k folds.\n",
|
||
"\n",
|
||
"L’objectif est d’obtenir une estimation plus robuste et plus stable de la performance du modèle, en particulier en présence de déséquilibre entre les classes. Le maintien de la structure du dataset initial dans chaque fold limite le risque de surapprentissage (overfitting) ou de sous-apprentissage sur certains folds dominés par une seule classe.\n",
|
||
"\n",
|
||
"3. Justification statistique\n",
|
||
"Le maintien de la distribution des classes dans les procédures d’échantillonnage est une exigence classique en statistique, relevant du principe de représentativité des échantillons. En classification supervisée, l'utilisation systématique de méthodes stratifiées permet d'améliorer la validité externe des résultats (capacité du modèle à généraliser) tout en réduisant la variance des estimations obtenues lors de la validation croisée."
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 20,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"name": "stderr",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"/Users/arthurdanjou/Workspace/studies/.venv/lib/python3.12/site-packages/keras/src/layers/core/dense.py:93: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
|
||
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
|
||
]
|
||
},
|
||
{
|
||
"name": "stdout",
|
||
"output_type": "stream",
|
||
"text": [
|
||
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 30ms/step\n",
|
||
" precision recall f1-score support\n",
|
||
"\n",
|
||
" Sain 0.71 0.45 0.56 11\n",
|
||
" Malade 0.65 0.85 0.73 13\n",
|
||
"\n",
|
||
" accuracy 0.67 24\n",
|
||
" macro avg 0.68 0.65 0.64 24\n",
|
||
"weighted avg 0.68 0.67 0.65 24\n",
|
||
"\n",
|
||
"0.7333333333333333\n"
|
||
]
|
||
}
|
||
],
|
||
"source": [
|
||
"from sklearn.model_selection import train_test_split\n",
|
||
"from sklearn.preprocessing import StandardScaler\n",
|
||
"from sklearn.metrics import f1_score, classification_report\n",
|
||
"import tensorflow as tf\n",
|
||
"import numpy as np\n",
|
||
"\n",
|
||
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42, stratify=y)\n",
|
||
"\n",
|
||
"scaler = StandardScaler()\n",
|
||
"X_train_scaled = scaler.fit_transform(X_train)\n",
|
||
"X_test_scaled = scaler.transform(X_test)\n",
|
||
"\n",
|
||
"model = build_model()\n",
|
||
"\n",
|
||
"model.compile(\n",
|
||
" optimizer='adam',\n",
|
||
" loss='binary_crossentropy'\n",
|
||
")\n",
|
||
"\n",
|
||
"callback = tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=10, restore_best_weights=True)\n",
|
||
"\n",
|
||
"history = model.fit(\n",
|
||
" X_train_scaled, y_train,\n",
|
||
" epochs=50,\n",
|
||
" batch_size=8,\n",
|
||
" validation_split=0.2,\n",
|
||
" callbacks=[callback],\n",
|
||
" verbose=0,\n",
|
||
" class_weight={0: 1.0, 1: 2.0}\n",
|
||
")\n",
|
||
"\n",
|
||
"\n",
|
||
"y_pred_test = (model.predict(X_test_scaled) > 0.5).astype(int)\n",
|
||
"\n",
|
||
"print(classification_report(y_test, y_pred_test, target_names=[\"Sain\", \"Malade\"]))\n",
|
||
"print(f1_score(y_test, y_pred_test))"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "code",
|
||
"execution_count": 21,
|
||
"metadata": {},
|
||
"outputs": [
|
||
{
|
||
"data": {
|
||
"image/png": "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",
|
||
"text/plain": [
|
||
"<Figure size 800x500 with 1 Axes>"
|
||
]
|
||
},
|
||
"metadata": {},
|
||
"output_type": "display_data"
|
||
}
|
||
],
|
||
"source": [
|
||
"plt.figure(figsize=(8, 5))\n",
|
||
"plt.plot(history.history['loss'], label='Loss (train)')\n",
|
||
"plt.plot(history.history['val_loss'], label='Loss (val)', linestyle='--')\n",
|
||
"plt.xlabel('Epochs')\n",
|
||
"plt.ylabel('Binary Cross-Entropy Loss')\n",
|
||
"plt.title('Courbe d\\'apprentissage')\n",
|
||
"plt.legend()\n",
|
||
"plt.grid(True)\n",
|
||
"plt.tight_layout()\n",
|
||
"plt.show()"
|
||
]
|
||
},
|
||
{
|
||
"cell_type": "markdown",
|
||
"metadata": {},
|
||
"source": [
|
||
"Lors de l'entraînement de mon réseau de neurones sur le dataset Breast Cancer Coimbra, j’ai obtenu un score F1 de 0.75, ce qui indique une bonne capacité du modèle à détecter les cas positifs (patients malades) tout en limitant les faux positifs.\n",
|
||
"\n",
|
||
"Un comportement particulier observé durant l’entraînement est que la val_loss est systématiquement inférieure à la train_loss. Ce phénomène s'explique principalement par l'utilisation de la régularisation L2, qui pénalise les poids uniquement pendant la phase d'entraînement, et non lors de l’évaluation sur les données de validation.\n",
|
||
"De plus, la taille réduite du dataset, l'emploi de class_weights pour compenser le léger déséquilibre des classes, ainsi que l'utilisation du early stopping, peuvent accentuer cet écart.\n",
|
||
"\n",
|
||
"Ce comportement n’est pas problématique tant que les performances en validation restent stables et satisfaisantes, ce qui est le cas ici avec un score F1 élevé, métrique prioritaire dans un contexte médical où le rappel est crucial."
|
||
]
|
||
}
|
||
],
|
||
"metadata": {
|
||
"kernelspec": {
|
||
"display_name": ".venv",
|
||
"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": "3.12.9"
|
||
}
|
||
},
|
||
"nbformat": 4,
|
||
"nbformat_minor": 2
|
||
}
|