{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"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","execution":{"iopub.status.busy":"2021-07-02T18:14:08.916445Z","iopub.execute_input":"2021-07-02T18:14:08.916953Z","iopub.status.idle":"2021-07-02T18:14:08.933636Z","shell.execute_reply.started":"2021-07-02T18:14:08.916835Z","shell.execute_reply":"2021-07-02T18:14:08.932318Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"!unzip ../input/dogs-vs-cats/train.zip -d train\n\n","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:14:59.573843Z","iopub.execute_input":"2021-07-02T18:14:59.574254Z","iopub.status.idle":"2021-07-02T18:15:11.874641Z","shell.execute_reply.started":"2021-07-02T18:14:59.574219Z","shell.execute_reply":"2021-07-02T18:15:11.873256Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os\nprint(os.listdir(\"../working\"))\n\n","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:17:08.734516Z","iopub.execute_input":"2021-07-02T18:17:08.734928Z","iopub.status.idle":"2021-07-02T18:17:08.742242Z","shell.execute_reply.started":"2021-07-02T18:17:08.73488Z","shell.execute_reply":"2021-07-02T18:17:08.74058Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom pathlib import Path\nimport os.path\nimport matplotlib.pyplot as plt\nimport tensorflow as tf\n\nfrom tqdm import tqdm\nfrom glob import glob\nimport pathlib\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nfrom PIL import Image\n\nfrom sklearn.model_selection import train_test_split\nimport matplotlib.image as mpimg\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom tensorflow.keras.optimizers import RMSprop\n\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import classification_report, confusion_matrix\n\n\n\nfrom tensorflow.keras.preprocessing import image\nfrom tensorflow.keras.applications import ResNet50\nfrom tensorflow.keras.applications import VGG16\nfrom tensorflow.keras.utils import to_categorical\nfrom tensorflow.keras.layers import GlobalAveragePooling2D, Dense, Flatten, Dropout\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom tensorflow.keras.layers import AveragePooling2D\nfrom tensorflow.keras.layers import Input\nfrom tensorflow.keras import Model\nfrom tensorflow.keras.models import Model\nfrom tensorflow.keras.optimizers import Adam\n\nfrom tensorflow.keras.utils import to_categorical\nfrom sklearn.preprocessing import LabelBinarizer\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import classification_report\nfrom sklearn.metrics import confusion_matrix\n\nimport warnings\nwarnings.filterwarnings(\"ignore\")","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:17:29.235193Z","iopub.execute_input":"2021-07-02T18:17:29.235744Z","iopub.status.idle":"2021-07-02T18:17:29.247384Z","shell.execute_reply.started":"2021-07-02T18:17:29.235706Z","shell.execute_reply":"2021-07-02T18:17:29.246221Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_csv_file = \"/kaggle/input/dogs-vs-cats/sampleSubmission.csv\"\ntrain_data = pd.read_csv(train_csv_file)\ntrain_data.sample(5)","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:17:49.770336Z","iopub.execute_input":"2021-07-02T18:17:49.770903Z","iopub.status.idle":"2021-07-02T18:17:49.788879Z","shell.execute_reply.started":"2021-07-02T18:17:49.770852Z","shell.execute_reply":"2021-07-02T18:17:49.787733Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Create a DataFrame including Filepath and Labels","metadata":{}},{"cell_type":"code","source":"def cat_convert(x):\n    x='cat.'+str(x)+'.jpg'\n    return x\ndef dog_convert(x):\n    x='dog.'+str(x)+'.jpg'\n    return x\n\ntrain_data['cat']=train_data['id'].apply(lambda x: cat_convert(x))\ntrain_data['dog']=train_data['id'].apply(lambda x: dog_convert(x))\n\ncat_label=pd.DataFrame(np.zeros(12500))\ndog_label=pd.DataFrame(np.ones(12500))\n\ntrain_label=cat_label.append(dog_label).reset_index(drop=True)\ntrain_label.shape","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:20.537933Z","iopub.execute_input":"2021-07-02T18:18:20.538424Z","iopub.status.idle":"2021-07-02T18:18:20.578595Z","shell.execute_reply.started":"2021-07-02T18:18:20.538385Z","shell.execute_reply":"2021-07-02T18:18:20.577187Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_label.columns=['Label']","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:23.330677Z","iopub.execute_input":"2021-07-02T18:18:23.331112Z","iopub.status.idle":"2021-07-02T18:18:23.338389Z","shell.execute_reply.started":"2021-07-02T18:18:23.331072Z","shell.execute_reply":"2021-07-02T18:18:23.336925Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_image=train_data['cat'].append(train_data['dog']).reset_index(drop=True)\n#train_image.columns=['image']\ntrain_image=pd.DataFrame({'image' : train_image})\ntrain_image","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:26.188796Z","iopub.execute_input":"2021-07-02T18:18:26.189242Z","iopub.status.idle":"2021-07-02T18:18:26.211235Z","shell.execute_reply.started":"2021-07-02T18:18:26.189204Z","shell.execute_reply":"2021-07-02T18:18:26.209875Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data=pd.DataFrame()\ndata['image']=train_image['image']\ndata['Label']=train_label['Label']","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:28.659915Z","iopub.execute_input":"2021-07-02T18:18:28.66038Z","iopub.status.idle":"2021-07-02T18:18:28.673091Z","shell.execute_reply.started":"2021-07-02T18:18:28.660341Z","shell.execute_reply":"2021-07-02T18:18:28.672153Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:31.075109Z","iopub.execute_input":"2021-07-02T18:18:31.075456Z","iopub.status.idle":"2021-07-02T18:18:31.091756Z","shell.execute_reply.started":"2021-07-02T18:18:31.075426Z","shell.execute_reply":"2021-07-02T18:18:31.090545Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"\n#image_path = \"C:\\\\Users\\\\niluf\\\\Desktop\\\\cats and dogs\\\\train\\\\train\\\\\"\nimage_path = '../working/train/train/'\n\nImage.open(image_path + data['image'][1000])","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:33.043841Z","iopub.execute_input":"2021-07-02T18:18:33.044265Z","iopub.status.idle":"2021-07-02T18:18:33.175923Z","shell.execute_reply.started":"2021-07-02T18:18:33.044223Z","shell.execute_reply":"2021-07-02T18:18:33.175117Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"Labelsdict={0: 'cat', 1 : ' dog'}\ndata['Labels'] =data['Label'].map(Labelsdict)\ndata.sample(5)","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:36.875845Z","iopub.execute_input":"2021-07-02T18:18:36.87639Z","iopub.status.idle":"2021-07-02T18:18:36.898935Z","shell.execute_reply.started":"2021-07-02T18:18:36.876355Z","shell.execute_reply":"2021-07-02T18:18:36.898102Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"image_paths = '../working/train/train/'\n\nnames=  data.image\nnames = pd.Series(names, name='Filepath').astype(str)\n#names = os.listdir(names)\nfor i in range(len(names)):\n    names[i] =  os.path.join(image_paths,names[i])\n    \nnames","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:40.466546Z","iopub.execute_input":"2021-07-02T18:18:40.466909Z","iopub.status.idle":"2021-07-02T18:18:40.885718Z","shell.execute_reply.started":"2021-07-02T18:18:40.466878Z","shell.execute_reply":"2021-07-02T18:18:40.884371Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"data['Filepath']=names","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:42.88707Z","iopub.execute_input":"2021-07-02T18:18:42.887466Z","iopub.status.idle":"2021-07-02T18:18:42.894184Z","shell.execute_reply.started":"2021-07-02T18:18:42.887431Z","shell.execute_reply":"2021-07-02T18:18:42.892711Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"final_data =data.drop(columns=['image'])\nfinal_data","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:46.694326Z","iopub.execute_input":"2021-07-02T18:18:46.694745Z","iopub.status.idle":"2021-07-02T18:18:46.718603Z","shell.execute_reply.started":"2021-07-02T18:18:46.694708Z","shell.execute_reply":"2021-07-02T18:18:46.717425Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# shuffle the DataFrame rows\nfinal_data = final_data.sample(frac = 1).reset_index(drop=True)\nfinal_data","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:18:49.310661Z","iopub.execute_input":"2021-07-02T18:18:49.31109Z","iopub.status.idle":"2021-07-02T18:18:49.332924Z","shell.execute_reply.started":"2021-07-02T18:18:49.31105Z","shell.execute_reply":"2021-07-02T18:18:49.331448Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# dividing data to train, validation and test datas","metadata":{}},{"cell_type":"code","source":"#train set to train and validation set\n#validation set model optimization\n\ntrain = final_data.iloc[0:18000,:]\nval= final_data.iloc[18000:22000:,:]\ntest =final_data.iloc[22000:,:]","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:31:23.766375Z","iopub.execute_input":"2021-07-02T18:31:23.766904Z","iopub.status.idle":"2021-07-02T18:31:23.774462Z","shell.execute_reply.started":"2021-07-02T18:31:23.766847Z","shell.execute_reply":"2021-07-02T18:31:23.773455Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Preprocessing images","metadata":{}},{"cell_type":"code","source":"train_generator = tf.keras.preprocessing.image.ImageDataGenerator(\n    preprocessing_function=tf.keras.applications.mobilenet_v2.preprocess_input,\n    validation_split=0.2\n)\n\ntest_generator = tf.keras.preprocessing.image.ImageDataGenerator(\n    preprocessing_function=tf.keras.applications.mobilenet_v2.preprocess_input\n)","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:31:27.614453Z","iopub.execute_input":"2021-07-02T18:31:27.615055Z","iopub.status.idle":"2021-07-02T18:31:27.620597Z","shell.execute_reply.started":"2021-07-02T18:31:27.615017Z","shell.execute_reply":"2021-07-02T18:31:27.619765Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_images = train_generator.flow_from_dataframe(\n    dataframe=train,\n    x_col='Filepath',\n    y_col='Labels',\n    target_size=(224, 224),\n    color_mode='rgb',\n    class_mode='categorical',\n    batch_size=32,\n    shuffle=True,\n    seed=0,\n    subset='training',\n    rotation_range=30,\n    zoom_range=0.15,\n    width_shift_range=0.2,\n    height_shift_range=0.2,\n    shear_range=0.15,\n    horizontal_flip=True,\n    fill_mode=\"nearest\"\n)\n\nval_images = train_generator.flow_from_dataframe(\n    dataframe=val,\n    x_col='Filepath',\n    y_col='Labels',\n    target_size=(224, 224),\n    color_mode='rgb',\n    class_mode='categorical',\n    batch_size=32,\n    shuffle=True,\n    seed=0,\n    subset='validation',\n    rotation_range=30,\n    zoom_range=0.15,\n    width_shift_range=0.2,\n    height_shift_range=0.2,\n    shear_range=0.15,\n    horizontal_flip=True,\n    fill_mode=\"nearest\"\n)\n\ntest_images = test_generator.flow_from_dataframe(\n    dataframe=test,\n    x_col='Filepath',\n    y_col='Labels',\n    target_size=(224, 224),\n    color_mode='rgb',\n    class_mode='categorical',\n    batch_size=32,\n    shuffle=True\n)","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:31:31.518046Z","iopub.execute_input":"2021-07-02T18:31:31.518675Z","iopub.status.idle":"2021-07-02T18:31:31.816452Z","shell.execute_reply.started":"2021-07-02T18:31:31.518618Z","shell.execute_reply":"2021-07-02T18:31:31.814984Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Model 1(Base_Model: VGG16)","metadata":{}},{"cell_type":"code","source":"\n\n\n#base_model = ResNet50( include_top=False, weights=None, input_shape=(224, 224, 3))\nbaseModel = VGG16(weights=\"imagenet\", include_top=False, input_tensor=Input(shape=(224, 224, 3)))\n\n# initialize the initial learning rate, number of epochs to train for,\n# and batch size\nINIT_LR = 1e-3\nEPOCHS = 10\nBS = 8\nheadModel = baseModel.output\nheadModel = AveragePooling2D(pool_size=(4, 4))(headModel)\nheadModel = Flatten(name=\"flatten\")(headModel)\nheadModel = Dense(64, activation=\"relu\")(headModel)\nheadModel = Dropout(0.5)(headModel)\nheadModel = Dense(2, activation=\"softmax\")(headModel)\n\n\n\nmodel = Model(inputs=baseModel.input, outputs=headModel)\n\nfor layer in baseModel.layers:\n    layer.trainable = False\n    \n\n# compile our model\nprint(\"[INFO] compiling model...\")\nopt = Adam(lr=INIT_LR, decay=INIT_LR / EPOCHS)\nmodel.compile(loss=\"binary_crossentropy\", optimizer=opt, metrics=[\"accuracy\"])\n\n\n\nhistory = model.fit(\n    train_images,\n    validation_data=val_images,\n    epochs=20,\n    callbacks=[\n        tf.keras.callbacks.EarlyStopping(\n            monitor='val_loss',\n            patience=5,\n            restore_best_weights=True\n        )\n    ]\n)\n","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.plot(history.history[\"loss\"], label=\"train_loss\")\nplt.plot(history.history[\"val_loss\"], label=\"val_loss\")\nplt.plot(history.history[\"acc\"], label=\"train_acc\")\nplt.plot(history.history[\"val_acc\"], label=\"val_acc\")\nplt.legend()\nplt.show()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Predict the label of the test_images\npred = model.predict(test_images)\npred = np.argmax(pred,axis=1)\n\npredict=pd.DataFrame({'pred':pred})\n\npredict['predict1']=predict['pred'].map(Labelsdict)\n\n","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score\ny_test = list(test.Labels)\npred1=predict['predict1']\nacc = accuracy_score(y_test,pred1)\nprint(f'Accuracy on the test set: {acc * 100:.2f}%')\n","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import classification_report\nclass_report = classification_report(y_test, pred1, zero_division=1)\nprint(class_report)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# # Model 2: Base Model: MobileNetV2","metadata":{}},{"cell_type":"code","source":"# Load the pretained model\npretrained_model = tf.keras.applications.MobileNetV2(\n    input_shape=(224, 224, 3),\n    include_top=False,\n    weights='imagenet',\n    pooling='max'\n)\n\npretrained_model.trainable = False","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:31:40.306144Z","iopub.execute_input":"2021-07-02T18:31:40.306518Z","iopub.status.idle":"2021-07-02T18:31:41.444303Z","shell.execute_reply.started":"2021-07-02T18:31:40.306487Z","shell.execute_reply":"2021-07-02T18:31:41.443017Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#inputs = pretrained_model.input\ninputs = pretrained_model.input\n\n\nx = tf.keras.layers.Dense(128, activation='relu')(pretrained_model.output)\nx = tf.keras.layers.Dense(128, activation='relu')(x)\n\noutputs = tf.keras.layers.Dense(2, activation='softmax')(x)\n\nmodel = tf.keras.Model(inputs=inputs, outputs=outputs)\n\nmodel.compile(\n    optimizer='adam',\n    loss='categorical_crossentropy',\n    metrics=['accuracy']\n)\n\nhistory = model.fit(\n    train_images,\n    validation_data=val_images,\n    epochs=10,\n    callbacks=[\n        tf.keras.callbacks.EarlyStopping(\n            monitor='val_loss',\n            patience=5,\n            restore_best_weights=True\n        )\n    ]\n)\n","metadata":{"execution":{"iopub.status.busy":"2021-07-02T18:31:47.623569Z","iopub.execute_input":"2021-07-02T18:31:47.624201Z","iopub.status.idle":"2021-07-02T19:26:05.027018Z","shell.execute_reply.started":"2021-07-02T18:31:47.624154Z","shell.execute_reply":"2021-07-02T19:26:05.025643Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\npd.DataFrame(history.history)[['accuracy','val_accuracy']].plot()\nplt.title(\"Accuracy\")\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2021-07-02T19:26:58.479264Z","iopub.execute_input":"2021-07-02T19:26:58.47973Z","iopub.status.idle":"2021-07-02T19:26:58.732805Z","shell.execute_reply.started":"2021-07-02T19:26:58.479689Z","shell.execute_reply":"2021-07-02T19:26:58.73136Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pd.DataFrame(history.history)[['loss','val_loss']].plot()\nplt.title(\"Loss\")\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2021-07-02T19:42:19.376097Z","iopub.execute_input":"2021-07-02T19:42:19.376785Z","iopub.status.idle":"2021-07-02T19:42:19.57702Z","shell.execute_reply.started":"2021-07-02T19:42:19.376741Z","shell.execute_reply":"2021-07-02T19:42:19.576251Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Predict the label of the test_images\npred = model.predict(test_images)\npred = np.argmax(pred,axis=1)\n\npredict=pd.DataFrame({'pred':pred})\n\npredict['predict1']=predict['pred'].map(Labelsdict)\n\n","metadata":{"execution":{"iopub.status.busy":"2021-07-02T19:42:27.70652Z","iopub.execute_input":"2021-07-02T19:42:27.707108Z","iopub.status.idle":"2021-07-02T19:43:56.249755Z","shell.execute_reply.started":"2021-07-02T19:42:27.707068Z","shell.execute_reply":"2021-07-02T19:43:56.248314Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test.shape","metadata":{"execution":{"iopub.status.busy":"2021-07-02T07:07:21.202158Z","iopub.execute_input":"2021-07-02T07:07:21.202568Z","iopub.status.idle":"2021-07-02T07:07:21.209612Z","shell.execute_reply.started":"2021-07-02T07:07:21.202535Z","shell.execute_reply":"2021-07-02T07:07:21.208514Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import accuracy_score\ny_test = list(test.Labels[:2999])\npred1=predict['predict1']\nacc = accuracy_score(y_test,pred1)\nprint(f'Accuracy on the test set: {acc * 100:.2f}%')\n","metadata":{"execution":{"iopub.status.busy":"2021-07-02T19:45:22.726219Z","iopub.execute_input":"2021-07-02T19:45:22.726831Z","iopub.status.idle":"2021-07-02T19:45:22.750902Z","shell.execute_reply.started":"2021-07-02T19:45:22.726793Z","shell.execute_reply":"2021-07-02T19:45:22.749489Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import classification_report\nclass_report = classification_report(y_test, pred1, zero_division=1)\nprint(class_report)","metadata":{"execution":{"iopub.status.busy":"2021-07-02T19:45:31.34452Z","iopub.execute_input":"2021-07-02T19:45:31.344914Z","iopub.status.idle":"2021-07-02T19:45:31.492553Z","shell.execute_reply.started":"2021-07-02T19:45:31.344882Z","shell.execute_reply":"2021-07-02T19:45:31.490912Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}