{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.11.11","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":3004,"databundleVersionId":861823,"sourceType":"competition"}],"dockerImageVersionId":31012,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\nfrom sklearn.model_selection import train_test_split\n\n#import sklearn\n#sklearn.set_config(transform_output=\"pandas\")\n\nimport tensorflow as tf\nimport tensorflow.keras.backend as K\nimport tensorflow.keras.layers as L\nimport tensorflow.keras.models as M\nimport tensorflow.keras.callbacks as C\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2025-04-18T07:22:02.379299Z","iopub.execute_input":"2025-04-18T07:22:02.379931Z","iopub.status.idle":"2025-04-18T07:22:02.387335Z","shell.execute_reply.started":"2025-04-18T07:22:02.379895Z","shell.execute_reply":"2025-04-18T07:22:02.386410Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df_train = pd.read_csv(\"/kaggle/input/digit-recognizer/train.csv\")\ndf_test = pd.read_csv(\"/kaggle/input/digit-recognizer/test.csv\")","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df_train.info()","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"LABEL = \"label\"\nFEATURES = [c for c in df_train.columns if c != LABEL]","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"y = df_train[LABEL]","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Transformer chaque ligne (image ) en matrice en 3 dimensions#\n\n# Utiliser la librairie numpy et les fonctions resize, reshape, concat\n\n# 1. Extraire les pixels\nX = df_train[FEATURES].values  # shape: (42000, 784)\n\n# 2. Normaliser les valeurs entre 0 et 1\nX = X / 255.0\n\n# 3. Reshaper chaque image en (28, 28, 1)\nX = X.reshape(-1, 28, 28, 1)  # -1 signifie \"autant de lignes que nécessaire\"\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-18T07:22:25.333865Z","iopub.execute_input":"2025-04-18T07:22:25.334559Z","iopub.status.idle":"2025-04-18T07:22:25.601664Z","shell.execute_reply.started":"2025-04-18T07:22:25.334525Z","shell.execute_reply":"2025-04-18T07:22:25.600955Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Préparer les données test (sans label)\nX_test = df_test.values / 255.0\nX_test = X_test.reshape(-1, 28, 28, 1)","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Faire train_test_split\n\n# Séparation en jeu d'entraînement et de validation\nX_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.1, random_state=42)\n\nprint(\"Train shape :\", X_train.shape, y_train.shape)\nprint(\"Validation shape :\", X_val.shape, y_val.shape)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-18T07:30:13.533248Z","iopub.execute_input":"2025-04-18T07:30:13.533926Z","iopub.status.idle":"2025-04-18T07:30:13.935859Z","shell.execute_reply.started":"2025-04-18T07:30:13.533900Z","shell.execute_reply":"2025-04-18T07:30:13.935028Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# TEster plusieurs modèles\n\nmodel = M.Sequential([\n    L.Conv2D(32, (3,3), padding='same', input_shape=(28,28,1)),\n    L.BatchNormalization(),\n    L.Activation('relu'),\n    L.MaxPooling2D(),\n    L.Dropout(0.2),\n\n    L.Conv2D(64, (3,3), padding='same'),\n    L.BatchNormalization(),\n    L.Activation('relu'),\n    L.MaxPooling2D(),\n    L.Dropout(0.2),\n\n    L.Conv2D(128, (3,3), padding='same'),\n    L.BatchNormalization(),\n    L.Activation('relu'),\n    L.MaxPooling2D(),\n    L.Dropout(0.2),\n\n    L.Flatten(),\n    L.Dense(128, activation='relu'),\n    L.Dropout(0.5),\n    L.Dense(10, activation='softmax')\n])\n\n\n# Compilation\nmodel.compile(optimizer='adam',\n              loss='sparse_categorical_crossentropy',\n              metrics=['accuracy'])\n\n# Entraînement\nmodel.fit(X_train, y_train, epochs=100, validation_data=(X_val, y_val))\n\n# Prédiction sur le test set\ny_pred = model.predict(X_test)\ny_pred_labels = np.argmax(y_pred, axis=1)\n\nmodel.summary()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-04-18T07:33:20.756058Z","iopub.execute_input":"2025-04-18T07:33:20.756345Z","iopub.status.idle":"2025-04-18T07:33:20.872492Z","shell.execute_reply.started":"2025-04-18T07:33:20.756326Z","shell.execute_reply":"2025-04-18T07:33:20.871687Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Création du fichier de soumission\nsubmission = pd.DataFrame({\n    \"ImageId\": np.arange(1, len(y_pred_labels) + 1),\n    \"Label\": y_pred_labels\n})\n\nsubmission.to_csv(\"/kaggle/working/submission.csv\", index=False)","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}