{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"from tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import Conv2D, BatchNormalization,DepthwiseConv2D,ZeroPadding2D,ReLU,GlobalAveragePooling2D, Dense\nimport numpy as np\nimport pandas as pd\nimport tensorflow.keras\nfrom PIL import Image, ImageOps\nfrom tensorflow.keras.layers import Concatenate,Input,Lambda\nimport tensorflow as tf\nfrom tensorflow.keras import layers\nimport keras\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras import initializers\nimport numpy as np\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras.models import Sequential\nfrom tensorflow.keras.layers import Activation, Dense, Flatten, BatchNormalization, Conv2D, MaxPool2D\nfrom tensorflow.keras.optimizers import Adam\nfrom tensorflow.keras.metrics import categorical_crossentropy\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom sklearn.metrics import confusion_matrix\nimport itertools\nimport os\nimport shutil\nimport random\nimport glob\nimport matplotlib.pyplot as plt\nimport warnings\n\n\nimport json\nimport math\nimport os\n\nimport cv2\nfrom PIL import Image\nimport numpy as np\nfrom keras import layers\nfrom keras.applications import DenseNet121\nfrom keras.callbacks import Callback, ModelCheckpoint\nfrom keras.preprocessing.image import ImageDataGenerator\nfrom keras.models import Sequential\nfrom keras.optimizers import Adam\nimport matplotlib.pyplot as plt\nimport pandas as pd\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import cohen_kappa_score, accuracy_score, auc, roc_auc_score, roc_curve\nimport sklearn\nimport scipy\nimport tensorflow as tf\nfrom tqdm import tqdm\nfrom keras.preprocessing import image\nfrom keras.models import Model\nfrom keras.layers import BatchNormalization, Dropout, Conv2D, MaxPooling2D, GlobalAveragePooling2D, Flatten, Dense\n\n%matplotlib inline\n\n\nwarnings.simplefilter(action='ignore', category=FutureWarning)\n#no zero pad\n#_URL = 'https://drive.google.com/file/d/1VZQe1rP0A7z4Xxa4omeyo89n7NgSaHXN/view?usp=sharing'","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"global batch_size\nglobal epochs\nglobal IMG_HEIGHT\nglobal IMG_WIDTH\nIMG_HEIGHT = 224\nIMG_WIDTH = 224\nnp.random.seed(2019)\ntf.random.set_seed(2019)\nTEST_SIZE = 0.25\nSEED = 2019\nBATCH_SIZE = 8","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_df = pd.read_csv('../input/aptos2019-blindness-detection/train.csv')\ntest_df = pd.read_csv('../input/aptos2019-blindness-detection/test.csv')\ntrain_df.head(7)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"x_train = np.load('../input/four-fold-aptos/train_all_four.npy')\nx_test = np.load('../input/four-fold-aptos/test_all_four.npy')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_train = train_df['diagnosis'].values\ny_train","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_train_one_hot = pd.get_dummies(train_df['diagnosis']).values\n\ny_train_multi = np.empty(y_train_one_hot.shape, dtype=y_train_one_hot.dtype)\ny_train_multi[:, 4] = y_train_one_hot[:, 4]\n\nfor i in range(3, -1, -1):\n    y_train_multi[:, i] = np.logical_or(y_train_one_hot[:, i], y_train_multi[:, i+1])\n\nprint(\"Original y_train:\", y_train_one_hot.sum(axis=0))\nprint(\"Multilabel version:\", y_train_multi.sum(axis=0))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# y_tr = y_train\nx_train, x_val, y_train, y_val = train_test_split(\n    x_train, y_train_multi, \n    test_size=TEST_SIZE, \n    random_state=SEED\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# train_image_generator = ImageDataGenerator(\n#                     rescale=1./255,\n#                     rotation_range=45,\n#                     width_shift_range=.15,\n#                     height_shift_range=.15,\n#                     horizontal_flip=True,\n#                     zoom_range=0.5\n#                     )\n#  # Generator for our training data\n# validation_image_generator =ImageDataGenerator(rescale=1./255)\n\n#  # Generator for our validation data\n# global train_data_gen\n# train_data_gen = train_image_generator.flow_from_directory(batch_size=batch_size,\n#                                                            directory=train_dir,\n#                                                            shuffle=True,\n#                                                            target_size=(IMG_HEIGHT, IMG_WIDTH),\n#                                                            class_mode='binary')\n# global val_data_gen\n# val_data_gen = validation_image_generator.flow_from_directory(batch_size=batch_size,\n#                                                               directory=validation_dir,\n#                                                               target_size=(IMG_HEIGHT, IMG_WIDTH),\n#                                                               class_mode='binary')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def create_datagen():\n    return ImageDataGenerator(\n        zoom_range=0.15,  # set range for random zoom\n        # set mode for filling points outside the input boundaries\n        fill_mode='constant',\n        cval=0.,  # value used for fill_mode = \"constant\"\n        horizontal_flip=True,  # randomly flip images\n        vertical_flip=True,  # randomly flip images\n    )\n\n# Using original generator\ndata_generator = create_datagen().flow(x_train, y_train, batch_size=BATCH_SIZE, seed=SEED)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# class string1(tensorflow.keras.layers.Layer):\n#     def __init__(self,filters1,filters2,chanDim=-1):\n#         super(string1, self).__init__()\n#         self.conv1s1=Conv2D(filters=filters1,kernel_size=(1,1),padding='same')\n#         self.bn1s1=BatchNormalization(axis=3,momentum=0.999)\n#         self.a1s1=ReLU(max_value=6,negative_slope=0,threshold=0)\n#         self.dcv1s1=DepthwiseConv2D(kernel_size=(3,3),padding='same')\n#         self.bn2s1=BatchNormalization(axis=3,momentum=0.999)\n#         self.a2s1=ReLU(max_value=6,negative_slope=0,threshold=0)\n#         self.conv2s1=Conv2D(filters=filters2,kernel_size=(1,1),padding='same')\n#         self.bn3s1=BatchNormalization(axis=3,momentum=0.999)\n\n#     def call(self,inputs):\n#         x=self.conv1s1(inputs)\n#         x=self.bn1s1(x)\n#         x=self.a1s1(x)\n#         x=self.dcv1s1(x)\n#         x=self.bn2s1(x)\n#         x=self.a2s1(x)\n#         x=self.conv2s1(x)\n#         x=self.bn3s1(x)\n#         return x\n\n# class string2(tensorflow.keras.layers.Layer):\n#     def __init__(self,filters1,filters2,chanDim=-1):\n#         super(string2, self).__init__()\n#         self.conv1s2=Conv2D(filters=filters1,kernel_size=(1,1),padding='same')\n#         self.bn1s2=BatchNormalization(axis=3,momentum=0.999)\n#         self.a1s2=ReLU(max_value=6,negative_slope=0,threshold=0)\n#         self.pads2=ZeroPadding2D(padding=([0,1],[0,1]),data_format='channels_last')\n#         self.dcv1s2=DepthwiseConv2D(kernel_size=(3,3),padding='same')\n#         self.bn2s2=BatchNormalization(axis=3,momentum=0.999)\n#         self.a2s2=ReLU(max_value=6,negative_slope=0,threshold=0)\n#         self.conv2s2=Conv2D(filters=filters2,kernel_size=(1,1),padding='same')\n#         self.bn3s2=BatchNormalization(axis=3,momentum=0.999)\n\n#     def call(self,inputs):\n#         x=self.conv1s2(inputs)\n#         x=self.bn1s2(x)\n#         x=self.a1s2(x)\n#         x=self.pads2(x)\n#         x=self.dcv1s2(x)\n#         x=self.bn2s2(x)\n#         x=self.a2s2(x)\n#         x=self.conv2s2(x)\n#         x=self.bn3s2(x)\n#         return x\n# #model=Sequential()\n\n# input1=Input(shape=(224,224,3))\n# zpd1=ZeroPadding2D(padding=([0,1],[0,1]),data_format='channels_last')(input1)\n# cv1=Conv2D(filters=16,kernel_size=(1,1),padding='same')(zpd1)\n# bn1=BatchNormalization(axis=3,momentum=0.999)(cv1)\n# a1=ReLU(max_value=6,negative_slope=0,threshold=0)(bn1)\n# dcv1=DepthwiseConv2D(kernel_size=(3,3),padding='same')(a1)\n# bn2=BatchNormalization(axis=3,momentum=0.999)(dcv1)\n# a2=ReLU(max_value=6,negative_slope=0,threshold=0)(bn2)\n# cv2=Conv2D(filters=8,kernel_size=(1,1),padding='same')(a2)\n# bn3=BatchNormalization(axis=3,momentum=0.999)(cv2)\n# cv3=Conv2D(filters=48,kernel_size=(1,1),padding='same')(bn3)\n# bn4=BatchNormalization(axis=3,momentum=0.999)(cv3)\n# a3=ReLU(max_value=6,negative_slope=0,threshold=0)(bn4)\n# zpd2=ZeroPadding2D(padding=([0,1],[0,1]),data_format='channels_last')(a3)\n# dcv2=DepthwiseConv2D(kernel_size=(3,3),padding='same')(zpd2)\n# bn5=BatchNormalization(axis=3,momentum=0.999)(dcv2)\n# a4=ReLU(max_value=6,negative_slope=0,threshold=0)(bn5)\n# cv4=Conv2D(filters=8,kernel_size=(1,1),padding='same')(a4)\n# bn6=BatchNormalization(axis=3,momentum=0.999)(cv4)\n# snz1=string1(48,8)(bn6)\n# merger1=Concatenate(axis=-1)([bn6,snz1])\n# sz1=string2(48,16)(merger1)\n# snz2=string1(96,16)(sz1)\n# merger2=Concatenate(axis=-1)([sz1,snz2])\n# snz3=string1(96,16)(merger2)\n# merger3=Concatenate(axis=-1)([merger2,snz3])\n# sz2=string2(96,24)(merger3)\n# snz4=string1(144,24)(sz2)\n# merger4=Concatenate(axis=-1)([sz2,snz4])\n# snz5=string1(144,24)(merger4)\n# merger5=Concatenate(axis=-1)([merger4,snz5])\n# snz6=string1(144,24)(merger5)\n# merger6=Concatenate(axis=-1)([merger5,snz6])\n# snz7=string1(144,32)(merger6)\n# snz8=string1(192,32)(snz7)\n# merger7=Concatenate(axis=-1)([snz7,snz8])\n# snz9=string1(192,32)(merger7)\n# merger8=Concatenate(axis=-1)([merger7,snz9])\n# sz3=string2(192,56)(merger8)\n# snz10=string1(336,56)(sz3)\n# merger9=Concatenate(axis=-1)([sz3,snz10])\n# snz11=string1(336,56)(merger9)\n# merger10=Concatenate(axis=-1)([merger9,snz11])\n# cv5=Conv2D(filters=336,kernel_size=(1,1),padding='same')(merger10)\n# bn7=BatchNormalization(axis=3,momentum=0.999)(cv5)\n# a5=ReLU(max_value=6,negative_slope=0,threshold=0)(bn7)\n# dcv3=DepthwiseConv2D(kernel_size=(3,3),padding='same')(a5)\n# bn8=BatchNormalization(axis=3,momentum=0.999)(dcv3)\n# a6=ReLU(max_value=6,negative_slope=0,threshold=0)(bn8)\n# cv7=Conv2D(filters=112,kernel_size=(1,1),padding='same')(a6)\n# bn9=BatchNormalization(axis=3,momentum=0.999)(cv7)\n# cv8=Conv2D(filters=1280,kernel_size=(1,1),padding='same')(bn9)\n# bn10=BatchNormalization(axis=3,momentum=0.999)(cv8)\n# a7=ReLU(max_value=6,negative_slope=0,threshold=0)(bn10)\n# pool1=GlobalAveragePooling2D(data_format='channels_last')(a7)\n# initializer=tf.keras.initializers.VarianceScaling(scale=1.0, mode=\"fan_in\", distribution=\"normal\", seed=None)\n# D1=Dense(units=100,activation='relu',batch_input_shape=(None,1280),kernel_initializer=initializer)(pool1)\n# D2=Dense(units=1,activation='softmax',use_bias='false',kernel_initializer=initializer)(D1)\n# model = tensorflow.keras.Model(inputs=input1, outputs=D2)\n\n# #keras.utils.plot_model(cmodel, \"my_first_model.png\")\n# model.compile(optimizer=Adam(lr=0.1),loss=tf.keras.losses.CategoricalCrossentropy(),metrics='accuracy')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = Sequential([\n    Conv2D(16, 3, padding='same', activation='relu', input_shape=(IMG_HEIGHT, IMG_WIDTH ,3)),\n    MaxPooling2D(),\n    Conv2D(32, 3, padding='same', activation='relu'),\n    MaxPooling2D(),\n    Conv2D(64, 3, padding='same', activation='relu'),\n    MaxPooling2D(),\n    BatchNormalization(axis=3,momentum=0.999),\n        ReLU(max_value=6,negative_slope=0,threshold=0),\n        Dropout(0.2),\n    DepthwiseConv2D(kernel_size=(3,3),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n    Conv2D(filters=8,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    Conv2D(filters=48,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n    ZeroPadding2D(padding=([0,1],[0,1]),data_format='channels_last'),\n    DepthwiseConv2D(kernel_size=(3,3),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n    Conv2D(filters=8,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    Conv2D(filters=192,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    ZeroPadding2D(padding=([0,1],[0,1]),data_format='channels_last'),\n    DepthwiseConv2D(kernel_size=(3,3),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n    Conv2D(filters=56,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n    Conv2D(filters=192,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n      Conv2D(filters=1024,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    ZeroPadding2D(padding=([0,1],[0,1]),data_format='channels_last'),\n    DepthwiseConv2D(kernel_size=(3,3),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Conv2D(filters=56,kernel_size=(1,1),padding='same'),\n    BatchNormalization(axis=3,momentum=0.999),\n    ReLU(max_value=6,negative_slope=0,threshold=0),\n    Dropout(0.2),\n    Flatten(),\n    Dense(512, activation='relu'),\n    Dense(5, activation='sigmoid')\n])\n\n# model.compile(optimizer=Adam(lr=0.0001),loss=tf.keras.losses.SquaredHinge(),metrics='accuracy')\nmodel.compile(loss='binary_crossentropy', optimizer=Adam(lr=0.00005), metrics=['accuracy','AUC'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.callbacks import EarlyStopping, ReduceLROnPlateau\n\nearly_stop = EarlyStopping(monitor='val_loss', min_delta=0.0001, patience=3, verbose=1, mode='auto')\n# Reducing the Learning Rate if result is not improving. \nreduce_lr = ReduceLROnPlateau(monitor='val_loss', min_delta=0.0004, patience=2, factor=0.1, min_lr=1e-6, mode='auto',\n                              verbose=1)\n\n# kappa_metrics = Metrics()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# history = model.fit_generator(\n#     train_data_gen,\n#         epochs=2,\n#     validation_data=val_data_gen,\n    \n# )\n\nhistory = model.fit_generator(\n    data_generator,\n    steps_per_epoch=x_train.shape[0] / BATCH_SIZE,\n    epochs=60,\n    validation_data=(x_val, y_val),\n    callbacks=[early_stop, reduce_lr]\n)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.save('net.h5')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_predict = model.predict(x_val)\ny_predict","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"val_y = y_predict > 0.5\nval_y = val_y.astype(int).sum(axis=1) - 1\nval_y","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_real = [4 if (list(i)[4]==1) else list(i).index(0)-1 for i in y_val]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"\n\nfrom sklearn.metrics import confusion_matrix \nfrom sklearn.metrics import accuracy_score \nfrom sklearn.metrics import classification_report \n  \nactual = y_real\npredicted = val_y\nresults = confusion_matrix(actual, predicted) \n  \nprint ('Confusion Matrix :')\nprint(results)\nprint ('Accuracy Score :',accuracy_score(actual, predicted) )\nprint ('Report : ')\nprint (classification_report(actual, predicted))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"predict_probab = [y_predict[i][val_y[i]] for i in range(len(val_y))]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_val_one_hot = []\nfor i in range(len(y_real)):\n    y_val_one_hot.append(list(np.zeros(5, dtype = 'uint8')))\ny_val_one_hot = np.array(y_val_one_hot)\nfor i in range(y_val_one_hot.shape[0]):\n    y_val_one_hot[i][y_real[i]] = 1","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def plot_roc(label):\n    y_probab = y_predict[:, label]\n    y_label = y_val_one_hot[:,label]\n    fpr, tpr, thresholds = roc_curve(y_label, y_probab)\n    auc = sklearn.metrics.auc(fpr, tpr)\n    plt.plot([0,1],[0,1], 'k--')\n    plt.plot(fpr,tpr, label = 'AUC SCORE : {:.3f}'.format(auc))\n    plt.title('AUC ROC Curve of class '+str(label))\n    plt.xlabel('False Positive rate')\n    plt.ylabel('True Positive rate')\n    plt.legend(loc = 'best')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_roc(0)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_roc(1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_roc(2)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_roc(3)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_roc(4)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.evaluate(x_val,y_val)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_test_p = model.predict(x_test)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_y = y_test_p > 0.5\ntest_y = test_y.astype(int).sum(axis=1) - 1\ntest_y","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"cohen_kappa_score(\n            y_real,\n            val_y, \n            weights='quadratic'\n        )","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"hist = history.history","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.figure(figsize=(8, 8))\nplt.title(\"Learning curve\")\nplt.plot(hist[\"loss\"], label=\"loss\")\nplt.plot(hist[\"val_loss\"], label=\"val_loss\")\n# plt.plot(np.argmin(hist[\"val_loss\"]), np.min(hist[\"val_loss\"]), marker=\"x\", color=\"r\",\n#          label=\"best model\")\nplt.xlabel(\"Epochs\")\nplt.ylabel(\"log_loss\")\nplt.legend();","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"\n\nplt.figure(figsize=(8, 8))\nplt.title(\"Learning curve\")\nplt.plot(hist[\"accuracy\"], label=\"accuracy\")\nplt.plot(hist[\"val_accuracy\"], label=\"val_acc\")\n# plt.plot(np.argmin(hist[\"val_loss\"]), np.min(hist[\"val_loss\"]), marker=\"x\", color=\"r\",\n#          label=\"best model\")\nplt.xlabel(\"Epochs\")\nplt.ylabel(\"accuracy\")\nplt.legend();","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.figure(figsize=(8, 8))\nplt.title(\"Learning curve\")\nplt.plot(hist[\"auc\"], label=\"loss\")\nplt.plot(hist[\"val_auc\"], label=\"val_loss\")\n# plt.plot(np.argmin(hist[\"val_loss\"]), np.min(hist[\"val_loss\"]), marker=\"x\", color=\"r\",\n#          label=\"best model\")\nplt.xlabel(\"Epochs\")\nplt.ylabel(\"auc\")\nplt.legend();","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}