{"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\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\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\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_count":null,"outputs":[]},{"cell_type":"code","source":"trainLabels = pd.read_csv(\"../input/aptos2019-blindness-detection/train.csv\")\ntrainLabels.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import numpy as np\nimport os\nimport time\nfrom keras.applications.vgg16 import VGG16\nfrom keras.preprocessing import image\nfrom keras.applications.vgg16 import preprocess_input\nfrom keras.layers import Dense, Activation, Flatten\nfrom keras.layers import merge, Input\nfrom keras.models import Model\nfrom keras.utils import np_utils\nfrom sklearn.utils import shuffle\nfrom sklearn.model_selection import train_test_split","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"base_model = VGG16(weights='imagenet', include_top=True, input_shape=(224, 224, 3))\n#freeze_layers(base_model)\nbase_model.summary()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"for layer in base_model.layers[:-3]:\n    layer.trainable = False\n# Check the trainable status of the individual layers\nfor layer in base_model.layers:\n    print(layer, layer.trainable)\nbase_model.summary()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.utils.vis_utils import plot_model\nplot_model(base_model, to_file='base_model_plot.png', show_shapes=True, show_layer_names=True)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from IPython.display import SVG\nfrom keras.utils.vis_utils import model_to_dot\n\nSVG(model_to_dot(base_model).create(prog='dot', format='svg'))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#batch_size to train\nbatch_size = 32\n# number of output classes\nnb_classes = 5\n# number of epochs to train\nnb_epoch = 10","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras import models\nfrom keras import layers\nfrom keras import optimizers\nfrom keras import regularizers\n \n# Create the model\nmodel = models.Sequential()\n \n# Add the vgg convolutional base model\nmodel.add(base_model)\n \n# Add new layers\nmodel.add(layers.Dense(1024, activation='relu'))\nmodel.add(layers.Dropout(0.5))\n# model.add(layers.Dense(4096, activation='relu'))\n# model.add(layers.Dropout(0.5))\n# model.add(Dense(500,kernel_regularizer=regularizers.l2(0.01),activation=\"relu\"))\nmodel.add(layers.Dense(nb_classes, activation='softmax', name ='output'))\n \n# Show a summary of the model. Check the number of trainable parameters\nmodel.summary()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.utils.vis_utils import plot_model\nplot_model(model, to_file='finetune_model_plot.png', show_shapes=True, show_layer_names=True)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from IPython.display import SVG\nfrom keras.utils.vis_utils import model_to_dot\n\nSVG(model_to_dot(model).create(prog='dot', format='svg'))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\ntrainLabels = pd.read_csv(\"../input/aptos2019-blindness-detection/train.csv\")\ntrainLabels.head()","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import seaborn as sns\nsns.countplot(\"diagnosis\",data= trainLabels)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os\n\nlisting = os.listdir(\"../input/aptos2019-blindness-detection/train_images/\") \n#listing.remove(\"train.csv\")\n#listing.remove(\"test.csv\")\nnp.size(listing)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from PIL import Image\n\n# input image dimensions\nimg_rows, img_cols = 224, 224\n\nimmatrix = []\nimlabel = []\n\nfor file in listing:\n    base = os.path.basename(\"../input/aptos2019-blindness-detection/train_images/\" + file)\n    fileName = os.path.splitext(base)[0]\n    imlabel.append(trainLabels.loc[(trainLabels.id_code)==fileName, 'diagnosis'].values[0])\n    im = Image.open(\"../input/aptos2019-blindness-detection/train_images/\" + file)   \n    img = im.resize((img_rows,img_cols))\n    rgb = img.convert('RGB')\n    immatrix.append(np.array(rgb).flatten())","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from sklearn.utils import shuffle\n\n#converting images & labels to numpy arrays\nimmatrix = np.asarray(immatrix)\nimlabel = np.asarray(imlabel)\n\n\ndata,Label = shuffle(immatrix,imlabel, random_state=2)\ntrain_data = [data,Label]\ntype(train_data)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport matplotlib\nfor i in range (10):\n    img=immatrix[i].reshape(img_rows,img_cols,3)\n    print('severity',imlabel[i])\n    if(imlabel[i]>0):\n        plt.imshow(img)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"(X, y) = (train_data[0],train_data[1])\nfrom sklearn.model_selection import train_test_split\n\n# STEP 1: split X and y into training and testing sets\n\nX_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=77)\nX_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size = 0.2, random_state = 77)\nX_train = X_train.reshape(X_train.shape[0], img_cols, img_rows, 3)\nX_test = X_test.reshape(X_test.shape[0], img_cols, img_rows, 3)\nX_val = X_val.reshape(X_val.shape[0], img_cols, img_rows, 3)\n\n\nprint(X_train.shape)\nprint(y_train.shape)\nprint(X_test.shape)\nprint(y_test.shape)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.preprocessing.image import ImageDataGenerator\nfrom keras.applications.vgg16 import preprocess_input\n\ntraining_datagen = ImageDataGenerator(\n                                    rescale=1./255,   # all pixel values will be between 0 an 1\n                                    shear_range=0.2, \n                                    zoom_range=0.2,\n                                    horizontal_flip=True,\n                                    preprocessing_function=preprocess_input)\n\nvalidation_datagen = ImageDataGenerator(rescale = 1./255, preprocessing_function=preprocess_input)\n\n","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# X_train = X_train.reshape(X_train.shape[0], img_cols, img_rows, 3)\n# X_test = X_test.reshape(X_test.shape[0], img_cols, img_rows, 3)\n\nX_train = X_train.astype('float32')\nX_test = X_test.astype('float32')\nX_val = X_val.astype('float32')\n\nX_train /= 255\nX_test /= 255\nX_val0 /= 255\n\nprint('X_train shape:', X_train.shape)\nprint(X_train.shape[0], 'train samples')\nprint(X_val.shape[0], 'val samples')\nprint(X_test.shape[0], 'test samples')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.utils import np_utils\n\n# convert class vectors to binary class matrices\nY_train = np_utils.to_categorical(y_train, nb_classes)\nY_val = np_utils.to_categorical(y_val, nb_classes)\nY_test = np_utils.to_categorical(y_test, nb_classes)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"from keras.preprocessing.image import ImageDataGenerator\n\n# create generators  - training data will be augmented images\nvalidationdatagenerator = ImageDataGenerator()\ntraindatagenerator = ImageDataGenerator(width_shift_range=0.1,height_shift_range=0.1,rotation_range=15,zoom_range=0.1 )\n\nbatchsize=12\ntrain_generator=traindatagenerator.flow(X_train, Y_train, batch_size=batchsize) \nvalidation_generator=validationdatagenerator.flow(X_val, Y_val,batch_size=batchsize)","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"model.compile(loss='categorical_crossentropy',\n              optimizer='adam',\n              metrics=['acc'])","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# summarize history for accuracy\nplt.plot(history.history['accuracy'])\nplt.plot(history.history['val_accuracy'])\nplt.title('model accuracy')\nplt.ylabel('accuracy')\nplt.xlabel('epoch')\nplt.legend(['train', 'test'], loc='upper left')\nplt.show()\nplt.savefig('model_accuracy.png')\n# summarize history for loss\nplt.plot(history.history['loss'])\nplt.plot(history.history['val_loss'])\nplt.title('model loss')\nplt.ylabel('loss')\nplt.xlabel('epoch')\nplt.legend(['train', 'test'], loc='upper left')\nplt.show()\nplt.savefig('model_loss.png')","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# from keras import backend as K\n# from keras.optimizers import RMSprop\n# K.set_value(model.optimizer.learning_rate, 0.00001)\n# train_datagen = ImageDataGenerator(rescale=1./255,\n#                                        shear_range=0.2,\n#                                        zoom_range=0.2,\n#                                        horizontal_flip=True)\n# test_datagen = ImageDataGenerator(rescale=1./255)\n# #train_batches = model.get_batches('train', gen=train_datagen, class_mode='categorical', batch_size=BATCH_SIZE)\n# #valid_batches = model.get_batches('valid', gen=test_datagen, class_mode='categorical', batch_size=BATCH_SIZE)\n# model.compile(optimizer=RMSprop(lr=0.00001), loss='categorical_crossentropy', metrics=['accuracy'])\n# #model.fit_generator(train_generator, epochs=15,\n#                            # steps_per_epoch=2000 // batchsize,\n#                            # validation_data=validation_generator,\n#                            # validation_steps=800 // batchsize)\n\nhistory= model.fit_generator(train_generator, steps_per_epoch=int(len(X_train)/batchsize), epochs=20, validation_data=validation_generator, validation_steps=int(len(X_test)/batchsize))","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"score = model.evaluate(X_test, Y_test, verbose=0)\nprint(score)","metadata":{"trusted":true},"execution_count":null,"outputs":[]}]}