image-classification-with-cnn.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":15444,"sourceType":"datasetVersion","datasetId":11102}],"dockerImageVersionId":30786,"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\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\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":"2024-11-17T13:28:18.578500Z","iopub.execute_input":"2024-11-17T13:28:18.579433Z","iopub.status.idle":"2024-11-17T13:28:18.595793Z","shell.execute_reply.started":"2024-11-17T13:28:18.579389Z","shell.execute_reply":"2024-11-17T13:28:18.594850Z"}},"outputs":[{"name":"stdout","text":"/kaggle/input/cifar10-python/cifar-10-python.tar.gz\n/kaggle/input/cifar10-python/cifar-10-batches-py/data_batch_1\n/kaggle/input/cifar10-python/cifar-10-batches-py/data_batch_2\n/kaggle/input/cifar10-python/cifar-10-batches-py/batches.meta\n/kaggle/input/cifar10-python/cifar-10-batches-py/test_batch\n/kaggle/input/cifar10-python/cifar-10-batches-py/data_batch_3\n/kaggle/input/cifar10-python/cifar-10-batches-py/data_batch_5\n/kaggle/input/cifar10-python/cifar-10-batches-py/data_batch_4\n/kaggle/input/cifar10-python/cifar-10-batches-py/readme.html\n","output_type":"stream"}],"execution_count":84},{"cell_type":"code","source":"import tarfile\nimport numpy as np\nimport pickle\nimport os\nimport cv2","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:18.597994Z","iopub.execute_input":"2024-11-17T13:28:18.598668Z","iopub.status.idle":"2024-11-17T13:28:18.602927Z","shell.execute_reply.started":"2024-11-17T13:28:18.598622Z","shell.execute_reply":"2024-11-17T13:28:18.602043Z"}},"outputs":[],"execution_count":85},{"cell_type":"code","source":"file_path = '/kaggle/input/cifar10-python/cifar-10-python.tar.gz' \nextracted_path = '/kaggle/working/cifar-10-batches-py'","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:18.604372Z","iopub.execute_input":"2024-11-17T13:28:18.604781Z","iopub.status.idle":"2024-11-17T13:28:18.611830Z","shell.execute_reply.started":"2024-11-17T13:28:18.604739Z","shell.execute_reply":"2024-11-17T13:28:18.610927Z"}},"outputs":[],"execution_count":86},{"cell_type":"code","source":"# Extraction du fichier tar.gz \nwith tarfile.open(file_path, 'r:gz') as tar: tar.extractall(path='/kaggle/working')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:18.613748Z","iopub.execute_input":"2024-11-17T13:28:18.614125Z","iopub.status.idle":"2024-11-17T13:28:20.180097Z","shell.execute_reply.started":"2024-11-17T13:28:18.614064Z","shell.execute_reply":"2024-11-17T13:28:20.178973Z"}},"outputs":[],"execution_count":87},{"cell_type":"code","source":"def load_cifar_batch(batch_filename): \n    with open(batch_filename, 'rb') as file: \n        batch = pickle.load(file, encoding='latin1') \n    data = batch['data'] \n    labels = batch['labels']\n    return data,labels","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:20.181465Z","iopub.execute_input":"2024-11-17T13:28:20.181790Z","iopub.status.idle":"2024-11-17T13:28:20.187056Z","shell.execute_reply.started":"2024-11-17T13:28:20.181757Z","shell.execute_reply":"2024-11-17T13:28:20.185922Z"}},"outputs":[],"execution_count":88},{"cell_type":"code","source":"# exemple\ndata,labels=load_cifar_batch(\"/kaggle/working/cifar-10-batches-py/data_batch_2\")\ndata[0]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:20.189788Z","iopub.execute_input":"2024-11-17T13:28:20.190156Z","iopub.status.idle":"2024-11-17T13:28:20.224130Z","shell.execute_reply.started":"2024-11-17T13:28:20.190120Z","shell.execute_reply":"2024-11-17T13:28:20.223253Z"}},"outputs":[{"execution_count":89,"output_type":"execute_result","data":{"text/plain":"array([ 35,  27,  25, ..., 169, 168, 168], dtype=uint8)"},"metadata":{}}],"execution_count":89},{"cell_type":"code","source":"data[0].shape\n# # Pour les données CIFAR-10, chaque image est initialement stockée sous forme aplatie en \n# une seule dimension, et toutes les images sont concaténées ensemble dans un tableau 2D \n# de forme (nombre_d_images, 3072)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:20.225609Z","iopub.execute_input":"2024-11-17T13:28:20.226031Z","iopub.status.idle":"2024-11-17T13:28:20.232417Z","shell.execute_reply.started":"2024-11-17T13:28:20.225983Z","shell.execute_reply":"2024-11-17T13:28:20.231515Z"}},"outputs":[{"execution_count":90,"output_type":"execute_result","data":{"text/plain":"(3072,)"},"metadata":{}}],"execution_count":90},{"cell_type":"code","source":"def redimensionner_et_gray(data, labels):\n    # Redimensionnement et transposition des données\n    data = data.reshape((-1, 3, 32, 32))  # Reshape en (nombre_d_images, canaux, hauteur, largeur)\n    data = data.transpose(0, 2, 3, 1)     # Transposition en (nombre_d_images, hauteur, largeur, canaux)\n\n    # Conversion des images en niveaux de gris et aplatissement en 1D\n    data_gray_1D = []\n    for img in data:\n        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)  # Conversion en niveaux de gris\n        gray_1D = gray.flatten()  # Aplatissement en 1D\n        data_gray_1D.append(gray_1D)\n\n    return np.array(data_gray_1D), labels","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:20.233825Z","iopub.execute_input":"2024-11-17T13:28:20.234164Z","iopub.status.idle":"2024-11-17T13:28:20.240487Z","shell.execute_reply.started":"2024-11-17T13:28:20.234124Z","shell.execute_reply":"2024-11-17T13:28:20.239643Z"}},"outputs":[],"execution_count":91},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\ndef show_examples(data, labels):\n    plt.figure(figsize=(6, 6))\n    for i in range(9):\n        plt.subplot(3, 3, i+1)\n        gray_image = data[i].reshape(32, 32)  # Remodeler les images en 32x32 pour l'affichage car on utilise flatten \n        plt.imshow(gray_image, cmap='gray', interpolation='nearest')\n        plt.title(labels[i])\n        plt.axis('off') \n    plt.tight_layout()\n    plt.show()\n\n            ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:20.241555Z","iopub.execute_input":"2024-11-17T13:28:20.241868Z","iopub.status.idle":"2024-11-17T13:28:20.252716Z","shell.execute_reply.started":"2024-11-17T13:28:20.241836Z","shell.execute_reply":"2024-11-17T13:28:20.251801Z"}},"outputs":[],"execution_count":92},{"cell_type":"code","source":"data,labels=load_cifar_batch(\"/kaggle/working/cifar-10-batches-py/data_batch_1\")\ndata,labels=redimensionner_et_gray(data,labels)\nshow_examples(data,labels)\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:20.253931Z","iopub.execute_input":"2024-11-17T13:28:20.254331Z","iopub.status.idle":"2024-11-17T13:28:21.136583Z","shell.execute_reply.started":"2024-11-17T13:28:20.254289Z","shell.execute_reply":"2024-11-17T13:28:21.135616Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 600x600 with 9 Axes>","image/png":"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"},"metadata":{}}],"execution_count":93},{"cell_type":"code","source":"data.shape","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.139841Z","iopub.execute_input":"2024-11-17T13:28:21.140182Z","iopub.status.idle":"2024-11-17T13:28:21.145886Z","shell.execute_reply.started":"2024-11-17T13:28:21.140149Z","shell.execute_reply":"2024-11-17T13:28:21.144945Z"}},"outputs":[{"execution_count":94,"output_type":"execute_result","data":{"text/plain":"(10000, 1024)"},"metadata":{}}],"execution_count":94},{"cell_type":"code","source":"sett=set(labels)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.147324Z","iopub.execute_input":"2024-11-17T13:28:21.147690Z","iopub.status.idle":"2024-11-17T13:28:21.153453Z","shell.execute_reply.started":"2024-11-17T13:28:21.147641Z","shell.execute_reply":"2024-11-17T13:28:21.152674Z"}},"outputs":[],"execution_count":95},{"cell_type":"code","source":"sett","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.154534Z","iopub.execute_input":"2024-11-17T13:28:21.154844Z","iopub.status.idle":"2024-11-17T13:28:21.163388Z","shell.execute_reply.started":"2024-11-17T13:28:21.154775Z","shell.execute_reply":"2024-11-17T13:28:21.162549Z"}},"outputs":[{"execution_count":96,"output_type":"execute_result","data":{"text/plain":"{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}"},"metadata":{}}],"execution_count":96},{"cell_type":"code","source":"#dans chaque batch afficher pour chaque classe nmbr de image pour chaque label\ndef countt(data, labels):\n    l = [0 for _ in range(10)]  # Initialisation de la liste de compteurs pour les 10 classes\n    for label in labels:\n        l[label] += 1\n    return l\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.164543Z","iopub.execute_input":"2024-11-17T13:28:21.164824Z","iopub.status.idle":"2024-11-17T13:28:21.170805Z","shell.execute_reply.started":"2024-11-17T13:28:21.164793Z","shell.execute_reply":"2024-11-17T13:28:21.169912Z"}},"outputs":[],"execution_count":97},{"cell_type":"code","source":"countt(data, labels)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.172060Z","iopub.execute_input":"2024-11-17T13:28:21.172390Z","iopub.status.idle":"2024-11-17T13:28:21.181834Z","shell.execute_reply.started":"2024-11-17T13:28:21.172346Z","shell.execute_reply":"2024-11-17T13:28:21.180912Z"}},"outputs":[{"execution_count":98,"output_type":"execute_result","data":{"text/plain":"[1005, 974, 1032, 1016, 999, 937, 1030, 1001, 1025, 981]"},"metadata":{}}],"execution_count":98},{"cell_type":"markdown","source":"**pour mon modéle je vais utiliser 2 batch**","metadata":{}},{"cell_type":"code","source":"\n# Charger les batches de données\nbatch1 = \"/kaggle/working/cifar-10-batches-py/data_batch_1\"\nbatch2 = \"/kaggle/working/cifar-10-batches-py/data_batch_2\"\nbatch3 = \"/kaggle/working/cifar-10-batches-py/data_batch_3\"\nbatch4 = \"/kaggle/working/cifar-10-batches-py/data_batch_4\"\n\n\ndata1, labels1 = load_cifar_batch(batch1)\ndata2, labels2 = load_cifar_batch(batch2)\ndata3, labels3 = load_cifar_batch(batch3)\ndata4, labels4 = load_cifar_batch(batch4)\n\n# Redimensionner et convertir les données en niveaux de gris 1D\ndata1, labels1 = redimensionner_et_gray(data1, labels1)\ndata2, labels2 = redimensionner_et_gray(data2, labels2)\ndata3, labels3 = redimensionner_et_gray(data3, labels3)\ndata4, labels4 = redimensionner_et_gray(data4, labels4)\n\n# Vérifier l'équilibre des labels dans les données\nprint(\"Nombre d'images dans le batch 1 pour chaque label:\", countt(data1, labels1))\nprint(\"Nombre d'images dans le batch 2 pour chaque label:\", countt(data2, labels2))\nprint(\"Nombre d'images dans le batch 3 pour chaque label:\", countt(data3, labels3))\nprint(\"Nombre d'images dans le batch 4 pour chaque label:\", countt(data4, labels4))\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.182988Z","iopub.execute_input":"2024-11-17T13:28:21.183371Z","iopub.status.idle":"2024-11-17T13:28:21.979629Z","shell.execute_reply.started":"2024-11-17T13:28:21.183330Z","shell.execute_reply":"2024-11-17T13:28:21.978585Z"}},"outputs":[{"name":"stdout","text":"Nombre d'images dans le batch 1 pour chaque label: [1005, 974, 1032, 1016, 999, 937, 1030, 1001, 1025, 981]\nNombre d'images dans le batch 2 pour chaque label: [984, 1007, 1010, 995, 1010, 988, 1008, 1026, 987, 985]\nNombre d'images dans le batch 3 pour chaque label: [994, 1042, 965, 997, 990, 1029, 978, 1015, 961, 1029]\nNombre d'images dans le batch 4 pour chaque label: [1003, 963, 1041, 976, 1004, 1021, 1004, 981, 1024, 983]\n","output_type":"stream"}],"execution_count":99},{"cell_type":"markdown","source":"**Voir les images pour chaque label**","metadata":{}},{"cell_type":"code","source":"def show_one_image_per_label(data, labels):\n    plt.figure(figsize=(10, 10)) \n    displayed_labels = [False] * 10 # Suivi des labels déjà affichés\n    count = 0 \n    for i in range(len(labels)):\n        label = labels[i]\n        if not displayed_labels[label]: # Vérifier si l'image de ce label a déjà été affichée \n            plt.subplot(3, 3, count + 1) \n            plt.imshow(data[i].reshape(32,32), interpolation='nearest',cmap='gray') \n            print(label)\n            plt.axis('off')\n            displayed_labels[label] = True \n            count += 1 \n        if count == 9: # Arrêter après avoir affiché 9 images\n            break \n        plt.tight_layout() \n        plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.980919Z","iopub.execute_input":"2024-11-17T13:28:21.981258Z","iopub.status.idle":"2024-11-17T13:28:21.988287Z","shell.execute_reply.started":"2024-11-17T13:28:21.981224Z","shell.execute_reply":"2024-11-17T13:28:21.987538Z"}},"outputs":[],"execution_count":100},{"cell_type":"code","source":"show_one_image_per_label(data2, labels2)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:21.989592Z","iopub.execute_input":"2024-11-17T13:28:21.989951Z","iopub.status.idle":"2024-11-17T13:28:22.403382Z","shell.execute_reply.started":"2024-11-17T13:28:21.989910Z","shell.execute_reply":"2024-11-17T13:28:22.402415Z"}},"outputs":[{"name":"stdout","text":"1\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 1000x1000 with 1 Axes>","image/png":"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"},"metadata":{}},{"name":"stdout","text":"6\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"name":"stdout","text":"8\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"name":"stdout","text":"3\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAJ8AAACgCAYAAAASN76YAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAANrElEQVR4nO2d2a5WRROGC1AQBRRxQkFAUBklRhlMlBhiIoGEG/AKOPdGvAMvgBBCAuFAhQRMAAVknkRw2AoCggOIiPKf0k+97tUSkwr53+esPrrXt9b6it5vV1dXj7lz586dMKaAsdU3YP5/sfOZMux8pgw7nynDzmfKsPOZMux8pgw7nynDzmfKeKC34fvvv9/YN27cSG1OnDjR2L/99ltjP/LII6nP/PnzG/vdd99NbZYuXdrYDz/8cGOfP38+9Tl06FBjf//994393HPPpT7PP/98Y48fPz61ISMjI439ww8/pDZ8xjVr1jS2WmT68MMPG3vTpk2pzdSpUxt79uzZjc13EBHxxx9/NPZjjz3W2Hy3ERHXr19v7DNnzqQ2Fy5caOzLly+nNsQjnynDzmfKsPOZMro136VLlxr71KlTqQ3/zj/55JONfe3atdRn69atjf3555+nNuvWrWvs9957r7Gp1dS9UBcqTfLUU0+lz8jNmzcb+/fffx/13yOyNhszZkxjb9myJfX5+eefG3vhwoWpDXX3Sy+91NhKv509e7axH3roocb+6quvUp8HHmjdhM8ToX/bITzymTLsfKYMO58pw85nyuiecJw+fbqxlcDkxIDBYRWc3Lt3b2PPmzcvtdm2bVtjT5gwobE3bNiQ+lCgHz16tLEZOI3IQfE///wztRk3blxj3759u7H//vvv1IeCnZM3BsAjItauXdvYDOBH5MkaJyBTpkxJfaZPn97Yr732WmNv3Lgx9Tl+/HhjMzAdETFp0qT02RAe+UwZdj5Thp3PlNGt+d55553G3rNnT2ozZ86c9uLQOtRqEREzZsxo7PXr16c21IGffvppY69evTr1YcCVmkQFma9cudLYPUFnopIErl692tgM7PK9RUT89ddfja001TPPPNPYTITgNSKyRmWA/u233059GHimNo6ImDhxYvpsCI98pgw7nynDzmfK6NZ8S5YsaeyehEIuQKuFbn6mNNOqVasam9ps3759qQ8TKx988MHGph5VbVSSAD+jreJ833777ahtVGIuP1N6efny5Y3NuKRKEuB7oU5ksmlE1okq/qlif0N45DNl2PlMGXY+U4adz5TRPeFgVvKyZctSm507dzY2A8hc1I7IgVKK/ogsgidPntzYXPiOyBMiBlz5PBE5w5gTm4iIixcvNva0adNGvbeInEjAHW5q990rr7zS2Cp7mJMmTmyeeOKJ1IeJBJysqQQGwgnIP302hEc+U4adz5Rh5zNldGu+AwcONPaLL76Y2lCbnTx5srGfffbZ1Ofpp59ubLUrjpqJO/FVYuuuXbsam4FdtRDOnWgq+KuqLtyN0j5MXP3yyy8b+7vvvkt9+K6UfqPme/311xtb6dqDBw82Nt/TL7/8kvpQuyuNqpINhvDIZ8qw85ky7HymDDufKaN7wrF9+/bGZgZLRMRbb73V2BTA586dS30YtFXZJtyFNXZs+39GZSXv3r27sRctWtTYPQJZ3QuzkIm6Licut27dauyffvop9eEzvvrqq6kNg+IMMnN3W0Se7PDeVFYRv0dlVbtchrmvsPOZMux8poxuzUcts3///tSG+m3x4sWNzSzaiBxkVsFUaqLPPvussVXGMbNtWXJM9aHGU5nXTBxgUFklI3D3GqseqF1yQ88ckYPvfEalw1Sm9d2oIDoDz7z/CB2QH8IjnynDzmfKsPOZMro1X0/1AcbbmIzAGJO6rloMf/TRRxubCac9VZMYA1MJAD2Vlni/jIEpqMWomdSxDNRmascYE2SpE6k1I/L98nlUHJM73FQscCj+qfDIZ8qw85ky7HymDDufKaN7wkFU+S0KUWYGc7E8IgdtVSYtxTcTDVSJMQpgBmTV/feU1BhCCW9OiBj8VROmxx9/vLFV8Jrw/TKBISJPFDnxUtnmDLar96LOnBvCI58pw85nyrDzmTK6RQ31G+2IrN+oAdViPvUOA8oRWZewjQoOD533qrQZ708lhg7pwJ6jEPjdSgvPnDmzsdVxCQw895x1R61L/czvjchBcXUvLotr7ivsfKYMO58po1vzUQuoDUTUZmoBmlDvqPgbF8ypLVnpKiLrnaEF9YhccYrHKURkfUmNp67LJFtW0FKJmLyOSvLkb0DNx0TdiFxFgnFVHkcWkWODhw8fTm0YR+3BI58pw85nyrDzmTLsfKaM7gkHBaXaZUaRz8mDytjl5EEFRllijOL7119/TX04EeCkRC2EM/P3xx9/HLxfinE1ETty5Ehjc/Felbw9duxY+oyoxIG7WbBgQfqMgWmWFP74449THwaZ1aSqJ6ObeOQzZdj5TBl2PlNGt+aj7uKOrIiIF154obHnzp3b2KqULoOcIyMjqQ0X/NlG7eyiTuF3M1kzIpfk/eSTTwbvhYFpdf+sSPDGG280ttKfDESrgD3vhUFl/h4R+XxdXkO9S97fvSQRKDzymTLsfKYMO58pw85nyuiecPAsMCWSGeylSB4Kiv5TG5UBPdr3RORgMHdg8QDpiDz5UWVl+Yy8NzWRWbduXWMzM0ZlXTPDW00E+N0M0Ks+LH3BgL1aCOB11Pl4KtA/hEc+U4adz5Rh5zNldGs+LhyrxAJqkLNnzza2ysZl8oHSbwwYM2Nala9l8JcBWKVtVqxY0dgsqxYR8dFHHzU2z5tjYF1dl8kH6l32VE/g7jomU/SU6OXuNVXyloFztVtQ6cshPPKZMux8pgw7nynjnhMLVDxr4cKFjU1tpmJ43LnF+FxE1hOqJC+hDmTCpvoetnnzzTdTm82bNzc2n3H16tWpDzUS9Zuq0sAzeKktI7KuZQKDir1Ru/NdfvPNN6kPd8X16M8ePPKZMux8pgw7nynDzmfK6J5wUKDPmjUrtWGwlAv158+f/9ffE5HLsfWU4eCEg8HUnrIWqg2fm+U+1KI7Jw9EnT3CjGOVOc6JCt9/TzD466+/bmxV9oKTSyYnROjfbQiPfKYMO58pw85nyrjnoxCOHj2aPlMl/YfgQnzP7ikGXNWOf94LtVnPojsTI9R1qAu5Ay4iJ0/w3lTQnEkYKkDP+6XGVrvMqJdVSTrCZ1ZVJZTWHbzuv+5hzH+Enc+UYeczZXRrvoMHD7YdRQyMO+RpKz3Hnfk9R071LMwzzsfjB1TSJDWUWmRnEio1njpTmLFB6i4Vt2RcUiUWMLbG8sGnT59OfagL+S5VlQMmKKhjMHq0I/HIZ8qw85ky7HymDDufKaN7wsEg54wZM1Ibingu1KvgJPuoMzWYRc2FbWYTq8+4s05VCeDkgeVsFT3n+DJJgAFZlXHMyZlKLOButaFnVm14DdWHkx+1Q48Txx488pky7HymDDufKWPMnZ7MzMi7stRCNxfDGfRUFQuYFKC0GDUS9aYKMlOX8LvVmbHUn2qxnPfHhE0VDOZ38Uw3lUzKxAell3lUA3cCqkoI1MsMOqvflcmw3CUXkStCfPDBB6kN8chnyrDzmTLsfKYMO58pozvITMGuMlSGdmlR3EZEnDt3rrEZ0IzIB00zUNqz46onSM5SZmr3F7OqmRWiJhwnTpwY1f7iiy9SH2aOqEnJUAaQOkSan7Fcm9qFxowg9Tur33YIj3ymDDufKcPOZ8ro1nysNsAzYyPy+WbUNmo3GLUY9UVE1nTULdxdFZF3iDFIvnz58tSHJceUrh1amO85I41n6e7bty/1GSqrFpEDu2yjKkQw8Mz3rZ6Zn6kgsyohPIRHPlOGnc+UYeczZXRrPu6EGjqSKiJrA9WHi+MqaZKxM8aUVMUCXnf//v2j32zksr4qsYAalQv+jFtG5ETL48ePN7bSc6w2oBIL+D5ZmlZVLGCc8uWXX25sVcmB70HtXlNafQiPfKYMO58pw85nyrDzmTK6M5mZAavEODOBOeFQEwMKaQZtI3Jgl2dJqNJsXCDnNaZPn576rFy5srFVYsHQLriTJ0+mPvxudVYx4SREZXgzuM5dciz/EZF/A05KVPYzg9XqvXAH3o4dO1Ib4pHPlGHnM2XY+UwZ3UFmaj5WEYjI5V2pU5ROpF5TAVfqLGootRiugrJ3w0BvRNZISg6zIgEDuyqBk7qKiRLq/ql9lV5m4q1KsCDUwtSf6l6YyKqe0UFmc19h5zNl2PlMGd2aj+VgR0ZGUhulF+5GbTKhVlAxO2omLmz3VGMasiNyZQG1e5/PSP2j9BC1Gb9bnVXLGCkTXdV1mBSgNh3xXfHdqngi9abS5ep9DuGRz5Rh5zNl2PlMGXY+U0b3hIMVAFTQk7vfKULVgjQnHEwaiMjBawpedXYZM6IZMFaTIwp/FUjnM/WU9eUkhMF2TuYi8gK/qoTAxXz+JirwOxT4VxUL2Edddyior/DIZ8qw85ky7HymjG7Nx7/pStuw3Cs1idrpPn/+/MZWZ6IxeMpyuz076agBlYaixlOJrdRiQ0mrETkozkA6z4mLyJpUVeJi8JpVJNR7oa6lflNanmfQqTY9Z+YRj3ymDDufKcPOZ8qw85kyulUiharKfqAYpzBVwdQlS5Y09uLFi1MbZpscOHBg1H+PyBk0zN5QpV05eVA7uZgpwkxgldXCiRYnHOrsNZ5bpwQ9ywVzktJz1l0PnIj1ZKT34JHPlGHnM2XY+UwZ3RULjPmv8chnyrDzmTLsfKYMO58pw85nyrDzmTLsfKYMO58pw85nyvgf/t4TW2LhQzoAAAAASUVORK5CYII="},"metadata":{}},{"name":"stdout","text":"4\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"name":"stdout","text":"0\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAJ8AAACgCAYAAAASN76YAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAIj0lEQVR4nO2dyYpUTRCFs51nHNqhFQdUVBBdKLhQEUGfQNC14MP5AiouFHQlKA4oouCA84Cz7Tz1v86TQd8wyaqo8v++XVTfvDfr9iE5GRWZOTIxMTGRAAKYEt0B+P+C+CAMxAdhID4IA/FBGIgPwkB8EAbigzAQH4QxzXvh+Ph45zUjIyNZPHXq1L/vkeO+w4b+iKTxnz9/+vKclMp36Wmj/bP6q+1GR0cn72xi5INAEB+EgfggDLfn8/i3KVNyLau/0L+n5PM7Xfe1GORiHf3OrTytfuffv38X1+izNLb+H/q/t/pb874Z+SAMxAdhID4IA/FBGO4JRyQ1hrxV4rYF1kSrixoDr9+55h79nKgx8kEYiA/CQHwQhtvzeXxXV1LZuofHD6kP8fSlVVFDC7oSuxaeH/z1M32XNV7TwuOfa3w5Ix+EgfggDMQHYTTN8/UrR+R5zrAXoPaKrmLSVm08MPJBGIgPwkB8EAbigzCaTjgw+e3olcmvoSZJ7oGRD8JAfBAG4oMwqgsLelWoGLkyLcqzelb19apvNX7O6i+r12CoQHwQBuKDMBAfhFE94fBUJatptqqLPROZFma75h6ttoXouq+nStlzn5pJYatrqGSGoQLxQRiID8Jwe74WK6Fa+bleeZtB+jHfQ82786yK6xeMfBAG4oMwEB+E0dfCAote5M1q79urvvSizb8AIx+EgfggDMQHYSA+CKPphKNrKy3r755t1Gq2SKuhV9XaXf21/u45w0Sp2QrY8xzrPI8WMPJBGIgPwkB8EIbb80WuGNNrfv78mcWfP38u2mh/p03Lv+rcuXM7n2P1rctX1Xwfj6fy+ELFc/aaB7ZIg38OxAdhID4Io3qXKk9OST2HJ4f369evzme9evUqi69cudL57GXLlmXxjh07ijbz588vPlN6kQt89OhRcc2XL1+yeMOGDcU1M2fO/Otnd9HPIgdGPggD8UEYiA/CQHwQRtMks04Mnj17lsU6UUipTP5aE45Pnz5l8evXr7P46tWrRZvx8fEsXr58eRbPmDGjaLN169YsXrBgQXFNi3MsPn78mMUXLlwo2rx//z6LlyxZUlyzYsWKLI48Y5gdC2CoQHwQBuKDMKqTzJ6tXNWr3bp1q2ij3sbyUFpIoM9R35hSStOnT5+0L+fOnSvaPH/+PIs3b95cXLN69eosnjNnThZ7toy9d+9eFj948KBoowlkq0hA34M+xyq40GctXLgwi1euXFm06dp9rBZGPggD8UEYiA/CQHwQRvXqNcsAq/HXSgxrW9w7d+5ksWW+dbJQU/mrVSJWklmf/eLFi+KajRs3ZvGmTZuyeGxsrGijEybrOypaYTN79uziGn0P+p5OnDhRtDl16lQW79u3L4uPHj1atNEJR4vt8lJi5INAEB+EgfggjKbn7WryUX2K+qOUyh/HV61aVVyjvvDSpUtZrAUMKZW+xJMM1v5ahQVv3rzJ4osXL2axJqFTKv2bejPLz82bNy+Lraplvc/169ez+OTJk0Wb8+fPZ/HevXuz2ErY96pggZEPwkB8EAbigzCqPZ9VAKB5PM21WXk+9VXbtm0rrlm3bl0Wa2GoesCUUrpx48akfVG/lFJKHz58yGItWk2pzF3++PEji9UTppTS2rVrs1j9m5VzVI/67t274hotoj179uykfUsppYMHD04ae3ZpIM8HQw/igzAQH4SB+CCMkQnnfg+6qsz6cV8nFGpMPefKelZBaRtrVdzly5ezWCcPL1++LNrcvXs3i3XSklI5QdL3snv37qLN4cOHs1jfy9OnT4s2mjjX1XhWf2/evJnF1oq3Y8eOZfH+/fuzuPbsX0UnTBaMfBAG4oMwEB+E0bSwoIuarV096PZnKaV04MCBLNaksuX5Tp8+ncXqqVJK6fHjx1l8+/btLLZWjGn/1GetWbOmaHPmzJksvnbtWnHN9+/fs1g9nhaKppTSzp07s7jFmb21MPJBGIgPwkB8EEa452txH8uDdP14v2jRoqKN/hBv+UL9MV8LCXShUkopHT9+PIt1G19ry9vR0dEs3rJlS3GNFoYuXbo0i/fs2VO00cJWj3/r1TEYjHwQBuKDMBAfhIH4IIy+Fhb0asJhoV/LswLr69evWWztLKCf6eq1+/fvF200Wa2FBNaKMZ0s7Nq1q7hGq5t1W7gjR44UbXR1YM279JwpbK3IUxj5IAzEB2EgPgijaZLZ4w1qqDl+oKZIVX2KldjVhLAen/Dw4cOije4koEUD1nbB6i016ZxS6bs1sf7kyZOijSbXXd7M4d1r/teMfBAG4oMwEB+EgfggjKZJ5prtUz2Pb7U9Qws0Wa3vQbfATak8a00nE1bFtE5SrAobfbZu42tVyxw6dCiL169fn8XWBETP6rC299D3wuo1GGgQH4SB+CCMnno+TUbWJic99+kXNQUL2l/1hd++fSvaaPGBlYhWf6Yr9Kxt4LZv357FixcvnrSv1jWWn1M94PlgoEF8EAbigzDwfH9JL4onrHepvlB3J0ip+33r/yylMkenntXaSlefo0Wr1md4PhhoEB+EgfggDMQHYfR1u4x/gZrJTlci2nMO3KxZs4prPOcOK10TJqtoQJ/DFmkw9CA+CAPxQRjVnm+QCjyHHc92wTU7C3hQP+rZjQDPB0MP4oMwEB+EQZ6vD7TwTJ5jqTyFrYPk1QenJ/C/A/FBGIgPwkB8EEZPJxyRFcfDjqdi2pMg7moTCSMfhIH4IAzEB2G4Pd/bt2+z2EpW6pb+eo1nRZOHXp0FVkNNX1oVCej79RSXdt239t3WeElGPggD8UEYiA/CQHwQhnvCoWc6WOa264DlsbGxoo2e12adQ2Z99q/Tr4RxzblqrWDkgzAQH4SB+CAM9xZpAK1h5IMwEB+EgfggDMQHYSA+CAPxQRiID8JAfBAG4oMw/gNit4aiwnxL4wAAAABJRU5ErkJggg=="},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"name":"stdout","text":"5\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"name":"stdout","text":"2\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAKAAAACgCAYAAACLz2ctAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAMaklEQVR4nO2dV49VxxKFi5yDAZNMsACZYIIAWQLNA4x/Dn+J38Abz1gIhIQQIBAZBpOsAQYYk4MD9/X21+uySyNLJXzX91aj7n367FPas6p2dfW0z58/fw5jiphevQDz/40d0JRiBzSl2AFNKXZAU4od0JRiBzSl2AFNKXZAU8rM7MAjR4409i+//NKNGR0dbez9+/c39uvXr7s5Hz58aOxPnz4NroVj/vjjj27M7NmzG3vmzPar/vnnn92cadOmDX72x48fG/vt27eN/e7du27OwoULG5vrV9+ZL6gWLFjQjZk+vX1+vH//Xqy45a+//vqiPWPGjMFrKDjv6NGjqXl+AppS7ICmFDugKSWtAanVaEdEzJ07t7HnzJnT2EpjccybN2+6MdRDvE5Gz1EfKd21cuXK7m9DUH9SI0b0+ojajXZEv/758+cPrkXdh6G1/P3334NryehENS+Dn4CmFDugKcUOaEqxA5pS0kEIE7tr167txqhk6ZeuEdEniJWIV3/70jUi+qCDQnrJkiXdHAppJeopwNVnk6EEd0bAq7VkPnsI9ZuQzPochJivEjugKcUOaEpJi4jFixc39sjISDeGGpDaLaMBlZZgspeaivouoi8SWL58eWMvWrSom6OuQ7i+jCbkmFmzZg1+DmHCPqK/D7y/SnvyN+F6MwUZaifvVHf3+gloSrEDmlLsgKaUtAZkHmrZsmXdmHnz5jU2X/grnUA9pPJd1IDUVNR7EX1hBPWRKmId0ncRvU7ky/yMhuIYpRu5FnVd6nLef3W/1b0agmv5J7u5+AloSrEDmlLsgKYUO6ApJR2EZF58U7RTrGYqdjNByKtXrxqb4luthcGCCkKIEtsMrHhdBT+Lol5VZzOIUmt58eLFF231m/G63LGXSZKrpHimYlvhJ6ApxQ5oSrEDmlLSGpA6S+kL6pShIoKIvpMA9V1EX1xKHaMKYfnSnWvJ6FFVnMB5mQIGzuG9zBRpPHnypBvz+PHjxn7+/PngdfkCgUUa33//fTeHuwWZAFfrzeInoCnFDmhKsQOaUuyAppS0cmSCUiVgKfQzbR8mJycb+/bt292YjRs3NjaFcyZ5zVYimQSyCjA4j4GXShhzfZkWaOPj443NACOi/45MBv/+++/dnGfPnjX2mjVrGpu/R0TErl27Glu1MGFgmMVPQFOKHdCUYgc0paQ14FSqhWlndrypF/NM3PKzlVYb2qmvihGo1ZROHKqAVnqU62OCmHovom9Tp4oE2PKYRQJKlw21QFYV0xn9PJWdfhF+Appi7ICmFDugKWXKGjDTDSlTBMpCg61bt3ZjmN9ioUFGq9FWWojryxQa8CW80sb8G/Wd0oCZbl7fffddY7OQQxUjrF+/vrFZWMCcX0T/m2SKP7L4CWhKsQOaUuyAphQ7oCklHYQw4ZppQcFARb3opkDnLq2IXuBOZS1Da4vIBRT8LCZ/1Q49VjNnCg2Gih4i+jbJmZ1pXO+KFSsaW1VEM+hQ93sqv0GEn4CmGDugKcUOaEpJa0AmbtX/fL6QZsJVvajnLi21u4p6iJ+jduozkcs5amcX15c5x5dacmJiopvD3WssIsgUYDx69KgbQ1goqvQ0Cwuo74aO2ojI/UZZ/AQ0pdgBTSl2QFNKWgOyaCCju9j14Jtvvunm8LgEle8aKoZVxZBcS2bTDL+jOhOZBQvUd2NjY4PX5WYhdUYytVim6JbXUTp96F6qAgbq0akWnyr8BDSl2AFNKXZAU4od0JSSDkL4oludtcYWsd2HiQQmX8Sr5OnSpUsbm0I5UzRAO9OiVyVXnz592tj3798fvC4T3Nx5pgI6Bk3qXJahNr5q/UMBnPqNOEYVcjgRbb5K7ICmFDugKSWtAVm4qGCClZrw7t273RzqudHR0W6M6sb032QKJDOaj4nnBw8edGP4N2pYpYVYfJDpjsUxSifyfnPnXOZcXyaZM7sdVVGJStpn8BPQlGIHNKXYAU0pdkBTSjoIuXLlSmOz0iWiF85sHcGEckS/s4stYxUMMDKtdDmHFSoRfXvgGzdudGO4s48VKKrSemgnnaqGyVSc8HszwMhUrQ+1PYkYbnPyv+Zl8BPQlGIHNKXYAU0paQ1IbbN69epuzKFDhxpbtRQj1DFKmzHxmWmjNtSO7eLFi92cs2fPNrYqruBaeB9UIpcFDCwiUIUdvHdKYw21zlUJY1al80WAatHL7+TOCOZfgx3QlGIHNKWkNeCmTZsae/Pmzd0Ydlai7lJnAQ8d7xXRa7xMXor65/r164196tSpbg5f7qviTObNuCtOFQCsWrXqi7YqlOB1VK6Q2pHrVTsB161b19gsRsisRZEpsFD4CWhKsQOaUuyAphQ7oCklHYSwKlkFIUxGUryqxGgmoBgqLFDBwr179xqbSWaemxvRB02qTQVhMlgl6Bl08LqqrW+mHd5QuxF+bkQfhDDAU4loogolVMV2Bj8BTSl2QFOKHdCUktaAW7ZsaWylW6jVVGEBybTFpc7iy3HqvYiIY8eONfb58+cbmzvVIvoksyqOZcKVuwWVFmJyN3PeMcm0l2NXCXXuHvUbCy6YjFeo4yBckGq+SuyAphQ7oCklrQHZnUnl65j3o7ZRL9Sp+ZSWoObj0VfHjx/v5pw7d66xqUdVLosbpJSe4/pUDpIMdY7K3EulWZmbPXjwYGPzbOCIXuO9fPmysZUGJyoneevWrcF5Cj8BTSl2QFOKHdCUYgc0paSDkOXLlze2ChZY8cwX/twdFtEHA6p7Aq974sSJxlZt3yjAGSyoDgYMdlTxBAMTFhao6xJeV53PxsBEjdm7d29jMynOau2IfrcdE9yqsplrUb/jzZs3u79l8BPQlGIHNKXYAU0paQ1IraaSp9QTTESrl+58sc1jDyJ6zUdts3v37m4OiyW4g0wde8AxaocYO2iNj48PzskU0BJqyx9++GHwuidPnmxsdb8PHDjQ2NSN6v7zO9KO0IUPGfwENKXYAU0pdkBTih3QlDLlA6vVuRBDZ6Cpqt5Lly419tWrV7sx3377bWOz6kO1NxtqVaZ2vA0lmSMitm/f3ti//vprY7MFSES/ft47lXy/cOFCYzPAiOiT3pnz5VhVw+BSVbXwXqozY9QZfxn8BDSl2AFNKXZAU8o/qgGZxOSOK1YpR/QaaseOHd2YnTt3NjaLBlSlNY+RYPGE2tVHDaU0IOf99NNPja0qramNmYhWuwc5h5XLEf13YqGBOmOPOpfXVd+ZWlK1IVY6NoOfgKYUO6ApxQ5oSklrwEwXq7GxscY+ffp0Y6tix3379jU2C18j+p1atNWxYdSFfDGvdnbxOuqYBmpAFlOoF/U88otaU3Xq4nXVDjfqRJ5lrDTgw4cPG5s5PaUBqfdVZzTmOrP4CWhKsQOaUuyAphQ7oCklHYQMJXYjIq5du9bYfPF9+PDhbg6vMzExMTiGQYiqMGaQxEpllfxlIYTa/cWWbUw8U+RH9EGGKp4gLDRQ35EBW6bNBs9w5n1Q939kZKSx2cIkQu8gzOAnoCnFDmhKsQOaUtIakElZlYj+8ccfG1sdE0DYLky1iGWRJBOsakcWNQlfoGfO31Vn27HwkhqQ5/FG9ElxdhHg0QkRubOWuT6e1cfz/SL6hPaZM2caWxVp8H6r3XaZ8+QUfgKaUuyAphQ7oCklrQGpmdT/fL6QHmoHq/6m8lDMJ7L4URU5sJsU16JyZMxvXbx4sRtDLUxNODk52c3hMRKco3JoXK/KHfLeUXOrYybY3pg6UR01xnyoKnLIHMmh8BPQlGIHNKXYAU0pdkBTSjoIobhWwpkFCxyjghDOUS1uGYSwPVtGALNaW1VEU4CrKl9WAzMA4udE9N+RQdSePXu6OQw6VNs0ftbGjRsbW539y6Q+u0yo78z7r9oFq92AGfwENKXYAU0pdkBTSloDUs+pzkvUBkxWqxfdmS4B27Zta2wmRn/77bduDrUZdaPq5sQ2uD///HM3hvDFvOoQwLOWqYVVp4HMmchDrXPVWnhfaCt9R12uEv/WgOarxA5oSrEDmlLsgKaUdBBy586dxmZLjYhewFK8qgoUVlaoFrFsgcaKGdX2jQKcAZFK0rLiRwUqly9fbmy20FCV4tyRx2pnFYSwbR1b9kb0gSEDLVZIR0Rs2LChsfkbqaob3hd171RLjwx+AppS7ICmFDugKWXa56luZzLmH8BPQFOKHdCUYgc0pdgBTSl2QFOKHdCUYgc0pdgBTSl2QFPKfwCzfZij7psSuAAAAABJRU5ErkJggg=="},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 0 Axes>"},"metadata":{}},{"name":"stdout","text":"7\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}}],"execution_count":101},{"cell_type":"code","source":"class_names = ['avion', 'automobile', 'oiseau', 'chat', 'cerf', 'chien', 'grenouille', 'cheval', 'bateau', 'camion']","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:22.404724Z","iopub.execute_input":"2024-11-17T13:28:22.405484Z","iopub.status.idle":"2024-11-17T13:28:22.411149Z","shell.execute_reply.started":"2024-11-17T13:28:22.405429Z","shell.execute_reply":"2024-11-17T13:28:22.409898Z"}},"outputs":[],"execution_count":102},{"cell_type":"code","source":"print(\"Valeurs avant normalisation (exemple):\", data1[0][:10]) # Affiche les 10 premières valeurs de la première image\nprint(\"min\",data1[0].min())\nprint(\"max\",data1[0].max())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:22.412626Z","iopub.execute_input":"2024-11-17T13:28:22.413110Z","iopub.status.idle":"2024-11-17T13:28:22.422024Z","shell.execute_reply.started":"2024-11-17T13:28:22.413043Z","shell.execute_reply":"2024-11-17T13:28:22.420923Z"}},"outputs":[{"name":"stdout","text":"Valeurs avant normalisation (exemple): [ 62  45  47  52  70  86 101 105 112 115]\nmin 0\nmax 247\n","output_type":"stream"}],"execution_count":103},{"cell_type":"code","source":"# Fonction pour normaliser une image\ndef normaliser(data): \n    # Normaliser les images\n    data = np.array([img/255.0 for img in data])\n    return data","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:22.423793Z","iopub.execute_input":"2024-11-17T13:28:22.424922Z","iopub.status.idle":"2024-11-17T13:28:22.430700Z","shell.execute_reply.started":"2024-11-17T13:28:22.424878Z","shell.execute_reply":"2024-11-17T13:28:22.429864Z"}},"outputs":[],"execution_count":104},{"cell_type":"code","source":"# Normaliser toutes les images dans data1 et data2\ndata2=normaliser(data2)\ndata1=normaliser(data1)\ndata3=normaliser(data3)\ndata4=normaliser(data4)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:22.431841Z","iopub.execute_input":"2024-11-17T13:28:22.432877Z","iopub.status.idle":"2024-11-17T13:28:22.773414Z","shell.execute_reply.started":"2024-11-17T13:28:22.432830Z","shell.execute_reply":"2024-11-17T13:28:22.772349Z"}},"outputs":[],"execution_count":105},{"cell_type":"code","source":"data1[0]\nprint(\"min\",data1[0].min())\nprint(\"max\",data1[0].max())","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:22.774952Z","iopub.execute_input":"2024-11-17T13:28:22.775280Z","iopub.status.idle":"2024-11-17T13:28:22.780237Z","shell.execute_reply.started":"2024-11-17T13:28:22.775247Z","shell.execute_reply":"2024-11-17T13:28:22.779348Z"}},"outputs":[{"name":"stdout","text":"min 0.0\nmax 0.9686274509803922\n","output_type":"stream"}],"execution_count":106},{"cell_type":"code","source":"show_examples(data1,labels1)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:22.781569Z","iopub.execute_input":"2024-11-17T13:28:22.781948Z","iopub.status.idle":"2024-11-17T13:28:23.427877Z","shell.execute_reply.started":"2024-11-17T13:28:22.781886Z","shell.execute_reply":"2024-11-17T13:28:23.426884Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 600x600 with 9 Axes>","image/png":"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"},"metadata":{}}],"execution_count":107},{"cell_type":"markdown","source":"**Model**","metadata":{}},{"cell_type":"code","source":"from keras.models import Model\nfrom keras.layers import Input, Conv2D, MaxPooling2D, Flatten, Dense, Dropout, BatchNormalization\nfrom keras.callbacks import EarlyStopping\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:28:23.429016Z","iopub.execute_input":"2024-11-17T13:28:23.429338Z","iopub.status.idle":"2024-11-17T13:28:23.434172Z","shell.execute_reply.started":"2024-11-17T13:28:23.429305Z","shell.execute_reply":"2024-11-17T13:28:23.433064Z"}},"outputs":[],"execution_count":108},{"cell_type":"code","source":"from keras.callbacks import EarlyStopping, ModelCheckpoint\nfrom keras.regularizers import l2\n\n# Définir le modèle CNN\ninput_img = Input(shape=(32, 32, 1))  # Note: Les images sont en niveaux de gris\n\n# Couches de convolution et de pooling\nx = Conv2D(32, (3, 3), activation='relu', padding='same', kernel_regularizer=l2(0.001))(input_img)\nx = BatchNormalization()(x)\nx = MaxPooling2D(pool_size=(2, 2))(x)\n\nx = Conv2D(64, (3, 3), activation='relu', padding='same', kernel_regularizer=l2(0.001))(x)\nx = BatchNormalization()(x)\nx = MaxPooling2D(pool_size=(2, 2))(x)\n\nx = Conv2D(128, (3, 3), activation='relu', padding='same', kernel_regularizer=l2(0.001))(x)\nx = BatchNormalization()(x)\nx = MaxPooling2D(pool_size=(2, 2))(x)\n\nx = Conv2D(256, (3, 3), activation='relu', padding='same', kernel_regularizer=l2(0.001))(x)\nx = BatchNormalization()(x)\nx = Flatten()(x)\n\n# Couches entièrement connectées avec Dropout\nx = Dense(512, activation='relu', kernel_regularizer=l2(0.001))(x) \nx = Dropout(0.5)(x)\nx = Dense(256, activation='relu', kernel_regularizer=l2(0.001))(x)\nx = Dropout(0.25)(x)\nx = Dense(128, activation='relu', kernel_regularizer=l2(0.001))(x)\nx = Dropout(0.5)(x)\noutput = Dense(10, activation='softmax')(x)  # 10 classes pour CIFAR-10\n\n# Définir et compiler le modèle\nmodel = Model(inputs=input_img, outputs=output)\nmodel.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:36:49.703872Z","iopub.execute_input":"2024-11-17T13:36:49.704735Z","iopub.status.idle":"2024-11-17T13:36:49.825566Z","shell.execute_reply.started":"2024-11-17T13:36:49.704693Z","shell.execute_reply":"2024-11-17T13:36:49.824535Z"}},"outputs":[],"execution_count":110},{"cell_type":"code","source":"from sklearn.model_selection import train_test_split\nimport numpy as np\n\n# Concaténer les quatre ensembles de données\ndata = np.concatenate((data1, data2, data3, data4), axis=0)\n\n# Concaténer les étiquettes correspondantes\nlabels = np.concatenate((labels1, labels2, labels3, labels4), axis=0)\n\nprint(f'Taille des données combinées: {data.shape}')\nprint(f'Taille des étiquettes combinées: {labels.shape}')\n\n# Adapter les dimensions des données pour correspondre aux entrées du modèle \ndata = data.reshape(data.shape[0], 32, 32, 1) # Diviser les données en ensembles d'entraînement et de validation \ndata_train, data_val, labels_train, labels_val = train_test_split( data, labels, test_size=0.2, random_state=42)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:36:54.694701Z","iopub.execute_input":"2024-11-17T13:36:54.695309Z","iopub.status.idle":"2024-11-17T13:36:54.922333Z","shell.execute_reply.started":"2024-11-17T13:36:54.695269Z","shell.execute_reply":"2024-11-17T13:36:54.921377Z"}},"outputs":[{"name":"stdout","text":"Taille des données combinées: (40000, 1024)\nTaille des étiquettes combinées: (40000,)\n","output_type":"stream"}],"execution_count":112},{"cell_type":"code","source":"# Ajout des callbacks Early Stopping et Model Checkpoint\nearly_stopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1, restore_best_weights=True)\ncheckpoint_filepath = '/kaggle/working/.{epoch:02d}-{val_loss:.2f}.keras'\nmodel_checkpoint = ModelCheckpoint(filepath=checkpoint_filepath, monitor='val_loss', verbose=1, save_best_only=True, mode='min')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:36:56.698498Z","iopub.execute_input":"2024-11-17T13:36:56.698924Z","iopub.status.idle":"2024-11-17T13:36:56.703940Z","shell.execute_reply.started":"2024-11-17T13:36:56.698871Z","shell.execute_reply":"2024-11-17T13:36:56.703035Z"}},"outputs":[],"execution_count":113},{"cell_type":"code","source":"from tensorflow.keras.preprocessing.image import ImageDataGenerator\n# Data Augmentation\ndatagen = ImageDataGenerator( rotation_range=20, \n                             width_shift_range=0.2,\n                             height_shift_range=0.2,\n                             horizontal_flip=True,\n                             zoom_range=0.2 )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:38:38.967148Z","iopub.execute_input":"2024-11-17T13:38:38.967543Z","iopub.status.idle":"2024-11-17T13:38:39.032618Z","shell.execute_reply.started":"2024-11-17T13:38:38.967506Z","shell.execute_reply":"2024-11-17T13:38:39.031597Z"}},"outputs":[],"execution_count":117},{"cell_type":"code","source":"# Adapter les données d'entraînement \ndatagen.fit(data_train)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:38:42.145634Z","iopub.execute_input":"2024-11-17T13:38:42.146375Z","iopub.status.idle":"2024-11-17T13:38:42.247522Z","shell.execute_reply.started":"2024-11-17T13:38:42.146335Z","shell.execute_reply":"2024-11-17T13:38:42.246691Z"}},"outputs":[],"execution_count":118},{"cell_type":"code","source":"\n# Entraîner le modèle en utilisant le générateur de données \nhistory = model.fit(datagen.flow(data_train, labels_train, batch_size=32),\n                    validation_data=(data_val, labels_val),\n                    epochs=150,\n                    callbacks=[early_stopping])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:38:43.013730Z","iopub.execute_input":"2024-11-17T13:38:43.014358Z","iopub.status.idle":"2024-11-17T13:48:32.074277Z","shell.execute_reply.started":"2024-11-17T13:38:43.014318Z","shell.execute_reply":"2024-11-17T13:48:32.073328Z"}},"outputs":[{"name":"stdout","text":"Epoch 1/150\n","output_type":"stream"},{"name":"stderr","text":"/opt/conda/lib/python3.10/site-packages/keras/src/trainers/data_adapters/py_dataset_adapter.py:121: UserWarning: Your `PyDataset` class should call `super().__init__(**kwargs)` in its constructor. `**kwargs` can include `workers`, `use_multiprocessing`, `max_queue_size`. Do not pass these arguments to `fit()`, as they will be ignored.\n  self._warn_if_super_not_called()\n","output_type":"stream"},{"name":"stdout","text":"\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m25s\u001b[0m 15ms/step - accuracy: 0.1624 - loss: 3.8969 - val_accuracy: 0.3215 - val_loss: 2.6032\nEpoch 2/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.2858 - loss: 2.6238 - val_accuracy: 0.2046 - val_loss: 3.3916\nEpoch 3/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.3595 - loss: 2.2917 - val_accuracy: 0.3264 - val_loss: 2.7402\nEpoch 4/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.4042 - loss: 2.1862 - val_accuracy: 0.4586 - val_loss: 2.0406\nEpoch 5/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.4507 - loss: 2.1325 - val_accuracy: 0.4956 - val_loss: 2.0340\nEpoch 6/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.4806 - loss: 2.0991 - val_accuracy: 0.4011 - val_loss: 2.6840\nEpoch 7/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.5008 - loss: 2.0644 - val_accuracy: 0.5879 - val_loss: 1.7780\nEpoch 8/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.5171 - loss: 2.0298 - val_accuracy: 0.5861 - val_loss: 1.7904\nEpoch 9/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.5342 - loss: 1.9672 - val_accuracy: 0.5784 - val_loss: 1.8080\nEpoch 10/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.5498 - loss: 1.9122 - val_accuracy: 0.5121 - val_loss: 2.1390\nEpoch 11/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 14ms/step - accuracy: 0.5554 - loss: 1.8970 - val_accuracy: 0.5224 - val_loss: 2.1633\nEpoch 12/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 14ms/step - accuracy: 0.5533 - loss: 1.8749 - val_accuracy: 0.5946 - val_loss: 1.8209\nEpoch 13/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5721 - loss: 1.8422 - val_accuracy: 0.5190 - val_loss: 2.0190\nEpoch 14/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5781 - loss: 1.8029 - val_accuracy: 0.5510 - val_loss: 1.8204\nEpoch 15/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5792 - loss: 1.7891 - val_accuracy: 0.6435 - val_loss: 1.5738\nEpoch 16/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5807 - loss: 1.7794 - val_accuracy: 0.4926 - val_loss: 2.0851\nEpoch 17/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5876 - loss: 1.7599 - val_accuracy: 0.5060 - val_loss: 2.1995\nEpoch 18/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5812 - loss: 1.7659 - val_accuracy: 0.6467 - val_loss: 1.5184\nEpoch 19/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 16ms/step - accuracy: 0.5910 - loss: 1.7299 - val_accuracy: 0.6010 - val_loss: 1.6612\nEpoch 20/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5863 - loss: 1.7373 - val_accuracy: 0.6187 - val_loss: 1.6793\nEpoch 21/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5957 - loss: 1.6896 - val_accuracy: 0.6300 - val_loss: 1.6039\nEpoch 22/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5969 - loss: 1.6995 - val_accuracy: 0.5771 - val_loss: 1.7455\nEpoch 23/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5968 - loss: 1.6864 - val_accuracy: 0.5261 - val_loss: 1.9696\nEpoch 24/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.5987 - loss: 1.6957 - val_accuracy: 0.5738 - val_loss: 1.8462\nEpoch 25/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6090 - loss: 1.6724 - val_accuracy: 0.6730 - val_loss: 1.4184\nEpoch 26/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6100 - loss: 1.6515 - val_accuracy: 0.6670 - val_loss: 1.4624\nEpoch 27/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6070 - loss: 1.6555 - val_accuracy: 0.6423 - val_loss: 1.5390\nEpoch 28/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6117 - loss: 1.6419 - val_accuracy: 0.6220 - val_loss: 1.6264\nEpoch 29/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6064 - loss: 1.6542 - val_accuracy: 0.6764 - val_loss: 1.3893\nEpoch 30/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6092 - loss: 1.6455 - val_accuracy: 0.6334 - val_loss: 1.5425\nEpoch 31/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6201 - loss: 1.6227 - val_accuracy: 0.5943 - val_loss: 1.6976\nEpoch 32/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6131 - loss: 1.6236 - val_accuracy: 0.5716 - val_loss: 1.7363\nEpoch 33/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6178 - loss: 1.6091 - val_accuracy: 0.6694 - val_loss: 1.4477\nEpoch 34/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6141 - loss: 1.6211 - val_accuracy: 0.5278 - val_loss: 1.9939\nEpoch 35/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6195 - loss: 1.6151 - val_accuracy: 0.6505 - val_loss: 1.4786\nEpoch 36/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6130 - loss: 1.6229 - val_accuracy: 0.5720 - val_loss: 1.8514\nEpoch 37/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6229 - loss: 1.5939 - val_accuracy: 0.6454 - val_loss: 1.4909\nEpoch 38/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6191 - loss: 1.6058 - val_accuracy: 0.6466 - val_loss: 1.5315\nEpoch 39/150\n\u001b[1m1000/1000\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m15s\u001b[0m 15ms/step - accuracy: 0.6199 - loss: 1.6093 - val_accuracy: 0.6531 - val_loss: 1.4523\nEpoch 39: early stopping\nRestoring model weights from the end of the best epoch: 29.\n","output_type":"stream"}],"execution_count":119},{"cell_type":"code","source":"# Évaluer le modèle\nloss, accuracy = model.evaluate(data_val, labels_val)\nprint(f'Loss: {loss}, Accuracy: {accuracy}')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:50:50.990802Z","iopub.execute_input":"2024-11-17T13:50:50.991492Z","iopub.status.idle":"2024-11-17T13:50:51.651702Z","shell.execute_reply.started":"2024-11-17T13:50:50.991439Z","shell.execute_reply":"2024-11-17T13:50:51.650696Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.6729 - loss: 1.3840\nLoss: 1.3892972469329834, Accuracy: 0.6763749718666077\n","output_type":"stream"}],"execution_count":120},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\n# Fonction pour tracer les graphes de précision et de perte\ndef plot_accuracy_loss(history):\n    # Tracer la précision d'entraînement et de validation\n    plt.figure(figsize=(12, 6))\n    \n    plt.subplot(1, 2, 1)\n    plt.plot(history.history['accuracy'], label='Train Accuracy')\n    plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n    plt.title('Accuracy over Epochs')\n    plt.xlabel('Epochs')\n    plt.ylabel('Accuracy')\n    plt.legend()\n    \n    # Tracer la perte d'entraînement et de validation\n    plt.subplot(1, 2, 2)\n    plt.plot(history.history['loss'], label='Train Loss')\n    plt.plot(history.history['val_loss'], label='Validation Loss')\n    plt.title('Loss over Epochs')\n    plt.xlabel('Epochs')\n    plt.ylabel('Loss')\n    plt.legend()\n    \n    plt.tight_layout()\n    plt.show()\n\n# Appeler la fonction pour afficher les graphes\nplot_accuracy_loss(history)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:50:59.349780Z","iopub.execute_input":"2024-11-17T13:50:59.350245Z","iopub.status.idle":"2024-11-17T13:51:00.461888Z","shell.execute_reply.started":"2024-11-17T13:50:59.350203Z","shell.execute_reply":"2024-11-17T13:51:00.460886Z"}},"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1200x600 with 2 Axes>","image/png":"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"},"metadata":{}}],"execution_count":121},{"cell_type":"markdown","source":"**faire meme etapes pour test images**","metadata":{}},{"cell_type":"code","source":"test_batch=\"/kaggle/working/cifar-10-batches-py/test_batch\"","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:51:15.506172Z","iopub.execute_input":"2024-11-17T13:51:15.507155Z","iopub.status.idle":"2024-11-17T13:51:15.511517Z","shell.execute_reply.started":"2024-11-17T13:51:15.507109Z","shell.execute_reply":"2024-11-17T13:51:15.510449Z"}},"outputs":[],"execution_count":122},{"cell_type":"code","source":"# Charger les batches de données de test\ndata_test, label_test = load_cifar_batch(test_batch)\n\n# Redimensionner et convertir les données en niveaux de gris 1D\ndata_test, label_test = redimensionner_et_gray(data_test, label_test)\n\n# Normaliser les données de test\ndata_test = normaliser(data_test)  # Assurez-vous de normaliser data_test\n\n# Redimensionner les données de test pour correspondre à l'entrée du modèle\ndata_test = data_test.reshape(data_test.shape[0], 32, 32, 1)\n\n# Assurez-vous que les étiquettes de test sont de type int\nlabel_test = np.array(label_test, dtype='int')\n\n# Évaluer le modèle sur les données de test\ntest_loss, test_accuracy = model.evaluate(data_test, label_test)\nprint(f'Test Loss: {test_loss}, Test Accuracy: {test_accuracy}')\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:51:16.528266Z","iopub.execute_input":"2024-11-17T13:51:16.529109Z","iopub.status.idle":"2024-11-17T13:51:18.872813Z","shell.execute_reply.started":"2024-11-17T13:51:16.529046Z","shell.execute_reply":"2024-11-17T13:51:18.871880Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 4ms/step - accuracy: 0.6717 - loss: 1.4070\nTest Loss: 1.4056476354599, Test Accuracy: 0.6711000204086304\n","output_type":"stream"}],"execution_count":123},{"cell_type":"code","source":"# Évaluer le modèle sur les données de test\ntest_loss, test_accuracy = model.evaluate(data_test,label_test)\nprint(f'Test Loss: {test_loss}, Test Accuracy: {test_accuracy}')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:51:20.251210Z","iopub.execute_input":"2024-11-17T13:51:20.251870Z","iopub.status.idle":"2024-11-17T13:51:21.112126Z","shell.execute_reply.started":"2024-11-17T13:51:20.251828Z","shell.execute_reply":"2024-11-17T13:51:21.110969Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.6717 - loss: 1.4070\nTest Loss: 1.4056476354599, Test Accuracy: 0.6711000204086304\n","output_type":"stream"}],"execution_count":124},{"cell_type":"markdown","source":"**Tester sur une image**","metadata":{}},{"cell_type":"code","source":"d, l = load_cifar_batch(\"/kaggle/working/cifar-10-batches-py/data_batch_3\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:51:23.796687Z","iopub.execute_input":"2024-11-17T13:51:23.797458Z","iopub.status.idle":"2024-11-17T13:51:23.827770Z","shell.execute_reply.started":"2024-11-17T13:51:23.797398Z","shell.execute_reply":"2024-11-17T13:51:23.826945Z"}},"outputs":[],"execution_count":125},{"cell_type":"code","source":"# Charger les données et les étiquettes du lot\ndata_batch_3, labels_batch_3 = load_cifar_batch(\"/kaggle/working/cifar-10-batches-py/data_batch_3\")\n\n# Choisir une image et l'étiquette associée\ntest_index =250 # Vous pouvez changer cet index pour tester une autre image\ntest_image = data_batch_3[test_index]\ntest_label = labels_batch_3[test_index]\n\n# Redimensionner et normaliser l'image\ntest_image = test_image.reshape(32, 32, 3)  # Reshape pour 32x32x3 (canaux RGB)\ntest_image = cv2.cvtColor(test_image, cv2.COLOR_RGB2GRAY)  # Convertir en niveaux de gris\ntest_image = cv2.resize(test_image, (32, 32))  # Assurez-vous que la taille est correcte\ntest_image = test_image.astype('float32') / 255.0  # Normaliser l'image\ntest_image = test_image.reshape(1, 32, 32, 1)  # Reshape pour correspondre à l'entrée du modèle\n\n# Assurez-vous que l'étiquette est un entier\ntest_label = np.array(test_label, dtype='int')\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:53:40.097589Z","iopub.execute_input":"2024-11-17T13:53:40.097982Z","iopub.status.idle":"2024-11-17T13:53:40.129175Z","shell.execute_reply.started":"2024-11-17T13:53:40.097945Z","shell.execute_reply":"2024-11-17T13:53:40.128065Z"}},"outputs":[],"execution_count":155},{"cell_type":"code","source":"# Faire une prédiction sur l'image de test\npredictions = model.predict(test_image)\npredicted_class = np.argmax(predictions)\n\n# Afficher les résultats\nclass_names = ['avion', 'automobile', 'oiseau', 'chat', 'cerf', 'chien', 'grenouille', 'cheval', 'bateau', 'camion']\nprint(f'Predicted class: {class_names[predicted_class]}')\nprint(f'Actual class: {class_names[test_label]}')\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:53:41.124888Z","iopub.execute_input":"2024-11-17T13:53:41.125631Z","iopub.status.idle":"2024-11-17T13:53:41.195031Z","shell.execute_reply.started":"2024-11-17T13:53:41.125589Z","shell.execute_reply":"2024-11-17T13:53:41.194138Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 18ms/step\nPredicted class: camion\nActual class: camion\n","output_type":"stream"}],"execution_count":156},{"cell_type":"code","source":"import numpy as np\nfrom sklearn.metrics import classification_report\n\n# Assurez-vous que data_test et label_test sont bien définis et prétraités\n# Exemple pour charger et prétraiter les données de test si nécessaire\ndata_test, label_test = load_cifar_batch(\"/kaggle/working/cifar-10-batches-py/test_batch\")\ndata_test, label_test = redimensionner_et_gray(data_test, label_test)\ndata_test = normaliser(data_test)\ndata_test = data_test.reshape(data_test.shape[0], 32, 32, 1)\nlabel_test = np.array(label_test, dtype='int')\n\n# Faire des prédictions sur les données de test\ny_pred = model.predict(data_test)\ny_pred_classes = np.argmax(y_pred, axis=1)  # Convertir les probabilités en classes prédictes\n\n# Générer le rapport de classification\nclass_names = ['avion', 'automobile', 'oiseau', 'chat', 'cerf', 'chien', 'grenouille', 'cheval', 'bateau', 'camion']\nreport = classification_report(label_test, y_pred_classes, target_names=class_names)\nprint(report)\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-11-17T13:53:14.131018Z","iopub.execute_input":"2024-11-17T13:53:14.131756Z","iopub.status.idle":"2024-11-17T13:53:15.860635Z","shell.execute_reply.started":"2024-11-17T13:53:14.131713Z","shell.execute_reply":"2024-11-17T13:53:15.859612Z"}},"outputs":[{"name":"stdout","text":"\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step\n              precision    recall  f1-score   support\n\n       avion       0.76      0.68      0.72      1000\n  automobile       0.75      0.86      0.80      1000\n      oiseau       0.74      0.35      0.48      1000\n        chat       0.48      0.46      0.47      1000\n        cerf       0.66      0.54      0.59      1000\n       chien       0.69      0.54      0.60      1000\n  grenouille       0.59      0.82      0.69      1000\n      cheval       0.62      0.82      0.71      1000\n      bateau       0.83      0.79      0.81      1000\n      camion       0.66      0.85      0.75      1000\n\n    accuracy                           0.67     10000\n   macro avg       0.68      0.67      0.66     10000\nweighted avg       0.68      0.67      0.66     10000\n\n","output_type":"stream"}],"execution_count":152},{"cell_type":"code","source":"","metadata":{"trusted":true},"outputs":[],"execution_count":null}]}