{"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":"2022-02-28T02:39:45.220511Z","iopub.execute_input":"2022-02-28T02:39:45.220764Z","iopub.status.idle":"2022-02-28T02:39:45.22904Z","shell.execute_reply.started":"2022-02-28T02:39:45.220735Z","shell.execute_reply":"2022-02-28T02:39:45.228148Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"######################################################Initial Packages########################################################\n#Basic Operating System Stuff\nimport os, gc, random\n\n#Basic dataframe, array, and math stuff\nimport pandas as pd #data frame\nimport math #math functions\nimport numpy as np    #numerical package\n\n#Scikit learn\nimport sklearn as sk  #scikit learn\nfrom sklearn import metrics\nfrom sklearn.model_selection import train_test_split as tts #train test split\nfrom sklearn.metrics import confusion_matrix, roc_curve, classification_report #for 2-class model\nfrom sklearn.preprocessing import MinMaxScaler as MMS\n\n#Tensorflow\nimport tensorflow as tf \nfrom tensorflow.python.client import device_lib #GPU Check\nimport tensorflow.keras #keras\nfrom tensorflow.keras import layers \nfrom tensorflow.keras import Sequential,Input,Model \nfrom tensorflow.keras.layers import Dense, Dropout, Flatten, Input, Add, Activation, ZeroPadding2D,GlobalAveragePooling2D\nfrom tensorflow.keras.layers import BatchNormalization,Conv2D, AveragePooling2D, MaxPooling2D, GlobalMaxPooling2D, LeakyReLU  \nfrom tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint #use for early stopping and reduction on level-out\nfrom tensorflow.keras.initializers import glorot_uniform, he_uniform #to initialize random weights for filters\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator \nfrom tensorflow.keras.preprocessing import image as image_utils\nfrom tensorflow.keras.applications.imagenet_utils import preprocess_input, decode_predictions, preprocess_input\nfrom tensorflow.keras.models import Model, load_model  #Can't do much without a model\nfrom tensorflow.keras import utils\nfrom tensorflow.keras.utils import get_file, plot_model, to_categorical, model_to_dot\nfrom tensorflow.keras.metrics import AUC\nfrom tensorflow.keras.preprocessing import image\nfrom tensorflow.keras.applications import ResNet50V2, MobileNetV2, VGG16 \nimport tensorflow.keras.backend as K #let's write our own metrics and loss functions\n\n#Graphing\nimport cv2\nfrom PIL import Image\nimport glob           #helps read files\nimport matplotlib.pyplot as plt\n\nprint(device_lib.list_local_devices()) #Let's see if Python recognizes my GPU, shall we?\n\nimport zipfile\n\nwith zipfile.ZipFile(\"../input/dogs-vs-cats-redux-kernels-edition/train.zip\",\"r\") as z:\n    z.extractall(\".\")\n    \nwith zipfile.ZipFile(\"../input/dogs-vs-cats-redux-kernels-edition/test.zip\",\"r\") as z:\n    z.extractall(\".\")\n\n# Any results you write to the current directory are saved as output.\n\n#print(os.listdir(\"/kaggle/working/train\"))","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:39:46.997768Z","iopub.execute_input":"2022-02-28T02:39:46.998315Z","iopub.status.idle":"2022-02-28T02:40:10.817907Z","shell.execute_reply.started":"2022-02-28T02:39:46.99828Z","shell.execute_reply":"2022-02-28T02:40:10.816709Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#Define function for clearing GPU memory\ndef reset_keras():\n    tensorflow.keras.backend.clear_session\n    gc.collect()\nreset_keras()","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:10.820363Z","iopub.execute_input":"2022-02-28T02:40:10.820794Z","iopub.status.idle":"2022-02-28T02:40:11.038127Z","shell.execute_reply.started":"2022-02-28T02:40:10.820752Z","shell.execute_reply":"2022-02-28T02:40:11.037277Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"filenames = os.listdir(\"./train\")\ncategories = []\nfor filename in filenames:\n    category = filename.split('.')[0]\n    if category == 'dog':\n        categories.append('dog')\n    else:\n        categories.append('cat')\n\ndf = pd.DataFrame({\n    'filename': filenames,\n    'category': categories\n})\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:11.040614Z","iopub.execute_input":"2022-02-28T02:40:11.041859Z","iopub.status.idle":"2022-02-28T02:40:11.090355Z","shell.execute_reply.started":"2022-02-28T02:40:11.041808Z","shell.execute_reply":"2022-02-28T02:40:11.089733Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"df['category'].value_counts().plot.bar()\ndf['category'].astype('str')","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:11.091919Z","iopub.execute_input":"2022-02-28T02:40:11.092161Z","iopub.status.idle":"2022-02-28T02:40:11.290389Z","shell.execute_reply.started":"2022-02-28T02:40:11.092127Z","shell.execute_reply":"2022-02-28T02:40:11.289701Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.preprocessing.image import ImageDataGenerator, load_img\nfrom sklearn.model_selection import train_test_split\nimport matplotlib.pyplot as plt\nimport random\n\nsample = random.choice(filenames)\nimage = load_img(\"./train/\"+sample)\nplt.imshow(image)\nprint(sample)","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:17.550783Z","iopub.execute_input":"2022-02-28T02:40:17.551464Z","iopub.status.idle":"2022-02-28T02:40:17.78718Z","shell.execute_reply.started":"2022-02-28T02:40:17.551427Z","shell.execute_reply":"2022-02-28T02:40:17.785664Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.models import Sequential\nfrom keras import layers\nfrom tensorflow.keras.optimizers import SGD\nfrom keras.layers import Conv2D, MaxPooling2D, Dropout, Flatten, Dense, Activation,GlobalMaxPooling2D\nfrom keras import applications\nfrom keras.preprocessing.image import ImageDataGenerator\nfrom keras import optimizers\nfrom tensorflow.keras.applications import VGG16\nfrom keras.models import Model\n\nimage_size = 224\ninput_shape = (image_size, image_size, 3)\n\nepochs = 5\nbatch_size = 16\n\npre_trained_model = VGG16(input_shape=input_shape, include_top=False, weights=\"imagenet\")\n    \nfor layer in pre_trained_model.layers[:15]:\n    layer.trainable = False\n\nfor layer in pre_trained_model.layers[15:]:\n    layer.trainable = True\n    \nlast_layer = pre_trained_model.get_layer('block5_pool')\nlast_output = last_layer.output\n    \n# Flatten the output layer to 1 dimension\nx = GlobalMaxPooling2D()(last_output)\n# Add a fully connected layer with 512 hidden units and ReLU activation\nx = Dense(512, activation='relu')(x)\n# Add a dropout rate of 0.5\nx = Dropout(0.5)(x)\n# Add a final sigmoid layer for classification\nx = layers.Dense(1, activation='sigmoid')(x)\n\nmodel = Model(pre_trained_model.input, x)\n\nmodel.compile(loss='binary_crossentropy',\n              optimizer=tensorflow.keras.optimizers.SGD(learning_rate=1e-4, momentum=0.9),\n              metrics=['accuracy'])\n\nmodel.summary()","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:19.859558Z","iopub.execute_input":"2022-02-28T02:40:19.860103Z","iopub.status.idle":"2022-02-28T02:40:20.942455Z","shell.execute_reply.started":"2022-02-28T02:40:19.860067Z","shell.execute_reply":"2022-02-28T02:40:20.941681Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_df, validate_df = train_test_split(df, test_size=0.1)\ntrain_df = train_df.reset_index()\nvalidate_df = validate_df.reset_index()\n\n# validate_df = validate_df.sample(n=100).reset_index() # use for fast testing code purpose\n# train_df = train_df.sample(n=1800).reset_index() # use for fast testing code purpose\n\ntotal_train = train_df.shape[0]\ntotal_validate = validate_df.shape[0]","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:27.827413Z","iopub.execute_input":"2022-02-28T02:40:27.828167Z","iopub.status.idle":"2022-02-28T02:40:27.838622Z","shell.execute_reply.started":"2022-02-28T02:40:27.828128Z","shell.execute_reply":"2022-02-28T02:40:27.837901Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_datagen = ImageDataGenerator(\n    rotation_range=15,\n    rescale=1./255,\n    shear_range=0.2,\n    zoom_range=0.2,\n    horizontal_flip=True,\n    fill_mode='nearest',\n    width_shift_range=0.1,\n    height_shift_range=0.1\n)\n\ntrain_generator = train_datagen.flow_from_dataframe(\n    train_df, \n    \"./train\", \n    x_col='filename',\n    y_col='category',\n    class_mode='binary',\n    target_size=(image_size, image_size),\n    batch_size=batch_size\n)","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:31.328315Z","iopub.execute_input":"2022-02-28T02:40:31.328813Z","iopub.status.idle":"2022-02-28T02:40:31.541692Z","shell.execute_reply.started":"2022-02-28T02:40:31.328776Z","shell.execute_reply":"2022-02-28T02:40:31.540813Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"validation_datagen = ImageDataGenerator(rescale=1./255)\nvalidation_generator = validation_datagen.flow_from_dataframe(\n    validate_df, \n    \"./train\", \n    x_col='filename',\n    y_col='category',\n    class_mode='binary',\n    target_size=(image_size, image_size),\n    batch_size=batch_size\n)","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:34.8936Z","iopub.execute_input":"2022-02-28T02:40:34.894163Z","iopub.status.idle":"2022-02-28T02:40:34.933319Z","shell.execute_reply.started":"2022-02-28T02:40:34.89412Z","shell.execute_reply":"2022-02-28T02:40:34.932582Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"example_df = train_df.sample(n=1).reset_index(drop=True)\nexample_generator = train_datagen.flow_from_dataframe(\n    example_df, \n    \"./train/\", \n    x_col='filename',\n    y_col='category'\n)\nplt.figure(figsize=(12, 12))\nfor i in range(0, 9):\n    plt.subplot(3, 3, i+1)\n    for X_batch, Y_batch in example_generator:\n        image = X_batch[0]\n        plt.imshow(image)\n        break\nplt.tight_layout()\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:37.682577Z","iopub.execute_input":"2022-02-28T02:40:37.682845Z","iopub.status.idle":"2022-02-28T02:40:39.145307Z","shell.execute_reply.started":"2022-02-28T02:40:37.682798Z","shell.execute_reply":"2022-02-28T02:40:39.143722Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# fine-tune the model\nhistory = model.fit_generator(\n    train_generator,\n    epochs=epochs,\n    validation_data=validation_generator,\n    validation_steps=total_validate//batch_size,\n    steps_per_epoch=total_train//batch_size)","metadata":{"execution":{"iopub.status.busy":"2022-02-28T02:40:43.900589Z","iopub.execute_input":"2022-02-28T02:40:43.901313Z","iopub.status.idle":"2022-02-28T03:06:28.538002Z","shell.execute_reply.started":"2022-02-28T02:40:43.901272Z","shell.execute_reply":"2022-02-28T03:06:28.537293Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.plot(history.history['accuracy'])\nplt.plot(history.history['val_accuracy'])\nplt.plot(history.history['loss'])\nplt.plot(history.history['val_loss'])\nplt.title('Model accuracy - 5 Epochs Optimal (Would Do More if More Time/Computing Power)')\nplt.ylabel('Accuracy')\nplt.xlabel('Epoch')\nplt.legend(['Train', 'Test', 'Loss', 'Test Loss'], loc='lower left')","metadata":{"execution":{"iopub.status.busy":"2022-02-28T03:10:51.097186Z","iopub.execute_input":"2022-02-28T03:10:51.097447Z","iopub.status.idle":"2022-02-28T03:10:51.334002Z","shell.execute_reply.started":"2022-02-28T03:10:51.097418Z","shell.execute_reply":"2022-02-28T03:10:51.333179Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.metrics import confusion_matrix, roc_curve, classification_report, auc #for 2-class model\ndef plt_roc_curve(name, model):\n    fpr, tpr, thresh = metrics.roc_curve(validation_generator.classes, model.predict(validation_generator))\n    x1=fpr\n    y1=tpr\n    auc = metrics.auc(x1,y1)\n    plt.plot(x1,y1,label='ROC curve for %s (AUC = %0.2f)' % (name, auc))\n    plt.plot([0,1], [0,1], 'k--')\n    plt.xlim((0,1.05))\n    plt.ylim((0,1.05))\n    plt.xlabel('FPR')\n    plt.ylabel('TPR')\n    plt.title('%s ROC Curve' % (name))\n    plt.legend(loc = 'lower right')\n    plt.show\n    \nplt_roc_curve(\"VGG16 - Seems Wrong?\", model)","metadata":{"execution":{"iopub.status.busy":"2022-02-28T03:15:43.69159Z","iopub.execute_input":"2022-02-28T03:15:43.691887Z","iopub.status.idle":"2022-02-28T03:15:51.225434Z","shell.execute_reply.started":"2022-02-28T03:15:43.691853Z","shell.execute_reply":"2022-02-28T03:15:51.224781Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import tensorflow\nfrom keras import layers\nfrom keras.layers import Conv2D, MaxPooling2D, Flatten, Dense, Dropout\nfrom keras.models import Model\nfrom keras import optimizers\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.optimizers import SGD\nmodel2 = Sequential()\ndef lenet_5(in_shape=(80,80,1), n_classes=1):\n    in_layer = layers.Input(in_shape)\n    conv1 = layers.Conv2D(filters=20, kernel_size=5,\n    padding='same', activation='relu')(in_layer)\n    pool1 = layers.MaxPool2D()(conv1)\n    conv2 = layers.Conv2D(filters=50, kernel_size=5,\n    padding='same', activation='relu')(pool1)\n    pool2 = layers.MaxPool2D()(conv2)\n    flatten = layers.Flatten()(pool2)\n    dense1 = layers.Dense(500, activation='relu',kernel_initializer='glorot_uniform')(flatten)\n    preds = layers.Dense(2, activation='softmax',kernel_initializer='glorot_uniform')(dense1)\n    opt = tensorflow.keras.optimizers.Adam(learning_rate=0.001, beta_1=0.9, beta_2=0.999, amsgrad=False)\n    model2 = Model(in_layer, preds)\n    model2.compile(loss=\"categorical_crossentropy\", optimizer=opt, metrics=[\"accuracy\"])\n    return model\nif __name__ == '__main__':\n    model2 = lenet_5()\nprint(model2.summary())","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"early_stopping_cb = keras.callbacks.EarlyStopping(patience=10)\nfrom keras.callbacks import ModelCheckpoint\ncheckpoint = ModelCheckpoint(\"lenet.h5\",verbose=1,save_best_only=True,monitor='val_accuracy', save_weights_only= False, mode ='auto',period=1)\nhistory2 = model2.fit(train_generator, validation_data=(validation_generator), batch_size=32, epochs=5, callbacks=[checkpoint, early_stopping_cb], shuffle=True)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.plot(history2.history['accuracy'])\nplt.plot(history2.history['val_accuracy'])\nplt.plot(history2.history['loss'])\nplt.plot(history2.history['val_loss'])\nplt.title('Model accuracy - 5 Epochs Optimal?')\nplt.ylabel('Accuracy')\nplt.xlabel('Epoch')\nplt.legend(['Train', 'Test', 'Loss', 'Test Loss'], loc='lower left')","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"p = {'first_hidden_layer': [300,500],\n'opt': [Adam, SGD],\n'dropout': [0,0.5],\n'weight_regulizer':[None],\n'learning_rate': [.001,.05],\n'emb_output_dims': [None],\n'kernel_initializer':[\"glorot_uniform\"]}","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pip install talos","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from tensorflow.keras.optimizers import Adam,SGD\nfrom keras.models import load_model\nfrom tensorflow.keras.utils import CustomObjectScope\nfrom tensorflow.keras.initializers import glorot_uniform\nimport talos\nfrom talos.model.normalizers import lr_normalizer\ndef lenet_model(x_train, y_train,x_val, y_val, params):\n    in_layer = layers.Input((80,80,1))\n    conv1 = layers.Conv2D(filters=20, kernel_size=5,\n    padding='same', activation='relu')(in_layer)\n    pool1 = layers.MaxPool2D()(conv1)\n    conv2 = layers.Conv2D(filters=50, kernel_size=5,\n    padding='same', activation='relu')(pool1)\n    pool2 = layers.MaxPool2D()(conv2)\n    flatten = layers.Flatten()(pool2)\n    dense1 = layers.Dense(params['first_hidden_layer'], activation='relu')(flatten)\n    dropout1 = layers.Dropout(params['dropout'])(dense1)\n    preds = layers.Dense(2, activation='softmax')(dropout1)\n    model = Model(in_layer, preds)\n    model.compile(loss=\"categorical_crossentropy\", optimizer=params['opt'](learning_rate=lr_normalizer(params['learning_rate'],params['opt'])), metrics=[\"acc\"])\n    steps_per_epoch = int(np.ceil(X_train.shape[0] / 20)) - 1\n    checkpoint3 = ModelCheckpoint(\"lenet.h5_2\",verbose=1,save_best_only=True,monitor='val_accuracy', save_weights_only= False, mode ='auto',period=1)\n    history2 = model.fit(train_generator, validation_data=(validation_generator), batch_size=64, epochs=7, callbacks=[talos.utils.ExperimentLogCallback('kgt', params), early_stopping_cb, checkpoint3], verbose=1)\n    return history2, model\nt = talos.Scan(train_generator, model=lenet_model, experiment_name= 'kgt', params=p)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import glob\nimport talos\npath = \"./kgt\"\nfilenames = glob.glob(path + \"/*.csv\")\nprint(filenames)\nr = talos.Reporting('./kgt/022722233347.csv')\nr.data","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"p = {'first_hidden_layer': [300],\n'opt': [Adam],\n'dropout': [0],\n'weight_regulizer':[None],\n'learning_rate': [.05],\n'emb_output_dims': [None],\n'kernel_initializer':[\"glorot_uniform\"]}\n\nmodel3 = Sequential()\ndef lenet_5_2(in_shape=(80,80,1), n_classes=1):\n    in_layer = layers.Input((80,80,1))\n    conv1 = layers.Conv2D(filters=20, kernel_size=5,\n    padding='same', activation='relu')(in_layer)\n    pool1 = layers.MaxPool2D()(conv1)\n    conv2 = layers.Conv2D(filters=20, kernel_size=5,\n    padding='same', activation='relu')(in_layer)\n    pool2 = layers.MaxPool2D()(conv1)\n    conv3 = layers.Conv2D(filters=50, kernel_size=5,\n    padding='same', activation='relu')(pool1)\n    pool3 = layers.MaxPool2D()(conv2)\n    flatten = layers.Flatten()(pool2)\n    dense1 = layers.Dense(params['first_hidden_layer'], activation='relu')(flatten)\n    dropout1 = layers.Dropout(params['dropout'])(dense1)\n    preds = layers.Dense(2, activation='softmax')(dropout1)\n    opt = tensorflow.keras.optimizers.Adam(learning_rate=0.05, kernel_initializer = 'glorot_uniform',\n                                         first_hidden_layer = 300)\n    model3 = Model(in_layer, preds)\n    model3.compile(loss=\"categorical_crossentropy\", optimizer=opt, metrics=[\"accuracy\"])\n    return model2\nif __name__ == '__main__':\n    model3 = lenet_5_2()\nprint(model3.summary())","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.callbacks import ModelCheckpoint\nearly_stopping_cb = keras.callbacks.EarlyStopping(patience=5)\ncheckpoint = ModelCheckpoint(\"lenet.h5_2\",verbose=1,save_best_only=True,monitor='val_accuracy', save_weights_only= False, mode ='auto',period=1)\nhistory3 = model3.fit(x=X_train_scaled, y=y_train_ohe, validation_data=(X_test_scaled, y_test_ohe), batch_size=32, epochs=4, callbacks=[checkpoint, early_stopping_cb], shuffle=True)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"plt.plot(history3.history['accuracy'])\nplt.plot(history3.history['val_accuracy'])\nplt.plot(history3.history['loss'])\nplt.plot(history3.history['val_loss'])\nplt.title('Model accuracy - 4 Epochs Optimal?')\nplt.ylabel('Accuracy')\nplt.xlabel('Epoch')\nplt.legend(['Train Accuracy', 'Test Accuracy', 'Train Loss', 'Test Loss'], loc='upper left')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_dir = \"test\"\npath = os.path.join(main_dir,train_dir)\n#os.listdir(path)\n\nX_test = []\nid_line = []\ndef create_test1_data(path):\n    for p in os.listdir(path):\n        id_line.append(p.split(\".\")[0])\n        img_array = cv2.imread(os.path.join(path,p),cv2.IMREAD_GRAYSCALE)\n        new_img_array = cv2.resize(img_array, dsize=(80, 80))\n        X_test.append(new_img_array)\ncreate_test1_data(path)\nX_test = np.array(X_test).reshape(-1,80,80,1)\nX_test = X_test/255","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"predict = model.predict_generator(test_generator, steps=np.ceil(nb_samples/batch_size))\nthreshold = 0.5\ntest_df['category'] = np.where(predict > threshold, 1,0)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"sample_test = test_df.sample(n=9).reset_index()\nsample_test.head()\nplt.figure(figsize=(12, 12))\nfor index, row in sample_test.iterrows():\n    filename = row['filename']\n    category = row['category']\n    img = load_img(\".test/\"+filename, target_size=(256, 256))\n    plt.subplot(3, 3, index+1)\n    plt.imshow(img)\n    plt.xlabel(filename + '(' + \"{}\".format(category) + ')')\nplt.tight_layout()\nplt.show()","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"submission_df = test_df.copy()\nsubmission_df['id'] = submission_df['filename'].str.split('.').str[0]\nsubmission_df['label'] = submission_df['category']\nsubmission_df.drop(['filename', 'category'], axis=1, inplace=True)\nsubmission_df.to_csv('submission_13010030.csv', index=False)\n\nplt.figure(figsize=(10,5))\nsns.countplot(submission_df['label'])\nplt.title(\"(Test data)\")","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def plt_roc_curve(name, model):\n    fpr, tpr, thresh = skplt.metrics.roc_curve(validation_data_gen.classes, model.predict(validation_data_gen))\n    x1=fpr\n    y1=fpr\n    auc skplt.metrics.auc(x1,y1)\n    plt.plot(x1,y1,label='ROC curve for %s (AUC = %0.2f)' % (name, auc))\n    plt.plot([0,1], [0,1], 'k--')\n    plt.xlim((0,1.05))\n    plt.ylim((0,1.05))\n    plt.xlabel('FPR')\n    plt.ylabel('TPR')\n    plt.title('%s ROC Curve' % (name))\n    plt.legend(loc = 'lower right')\n    plt.show","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"predictions = model.predict(X_test)\npredictions2 = model2.predict(X_test)\npredictions3 = model3.predict(X_test)\n\npredicted_val = [int(round(p[0])) for p in predictions]\npredicted_val2 = [int(round(p[0])) for p in predictions2]\npredicted_val3 = [int(round(p[0])) for p in predictions3]\n\nsubmission_df = pd.DataFrame({'id':id_line, 'label':predicted_val})\nsubmission_df.to_csv(\"submission.csv\", index=False)\n\nsubmission_df = pd.DataFrame({'id':id_line, 'label':predicted_val2})\nsubmission_df.to_csv(\"submission2.csv\", index=False)\n\nsubmission_df = pd.DataFrame({'id':id_line, 'label':predicted_val3})\nsubmission_df.to_csv(\"submission3.csv\", index=False)","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}