{"cells":[{"metadata":{},"cell_type":"markdown","source":"In this kernel I would use Keras to build a baseline, this type of baseline can be helpful to you in solving similar problems as well."},{"metadata":{},"cell_type":"markdown","source":"### Update 2 (13/12):\nTrying out Noisy Student weights\n\n### Update 1 (11/12):\n\nI am removing normalization step in generator since in EfficientNet, normalization is done within the model itself and the model expects input in the range of [0,255]"},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import pandas as pd\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport tensorflow as tf\nfrom tensorflow.keras.models import Sequential, Model\nfrom tensorflow.keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau\nfrom tensorflow.keras.layers import Dense, Dropout, Flatten,GlobalAveragePooling2D,BatchNormalization, Activation\nimport glob\n\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator\nfrom sklearn.model_selection import train_test_split\nfrom tensorflow import keras\n\nimport os","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Adding Seed helps to reproduce results. Setting Debug Parameter will run the model on smaller number of epochs to validate the architecture."},{"metadata":{},"cell_type":"markdown","source":"## Prepare Data"},{"metadata":{"trusted":true},"cell_type":"code","source":"SEED = 42\nDEBUG = False","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df = pd.read_csv('../input/cassava-leaf-disease-classification/train.csv')\ndf.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Distribution of dataset:"},{"metadata":{"trusted":true},"cell_type":"code","source":"df['path'] = '../input/cassava-leaf-disease-classification/train_images/' + df['image_id']\ndf.label.value_counts(normalize=True) * 100\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Reading Test Images\n\ntest_images = glob.glob('../input/cassava-leaf-disease-classification/test_images/*.jpg')\ndf_test = pd.DataFrame(test_images, columns = ['path'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"if DEBUG:\n    _, df = train_test_split(df, test_size = 0.1, random_state=SEED, shuffle=True, stratify=df['label'])\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Augmentation"},{"metadata":{"trusted":true},"cell_type":"code","source":"def my_image_augmentation(train=True):\n    if train:\n        my_idg = ImageDataGenerator(#rescale=1. / 255.0,\n                                    horizontal_flip = True, \n                                    vertical_flip = True, \n                                    height_shift_range=0.2, \n                                    width_shift_range=0.2, \n                                    brightness_range=[0.7, 1.5],\n                                    rotation_range=30, \n                                    shear_range=0.2,\n                                    fill_mode='nearest',\n                                    zoom_range=[0.3,0.6],\n            \n            #featurewise_center=True, samplewise_center=True,\n        )\n    else:\n        #my_idg = ImageDataGenerator(#rescale=1. / 255.0) # No transformations on the validation/test set\n        my_idg = ImageDataGenerator()\n    \n    return my_idg\n\n\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"Once we have the Generator we will feed in the data to generator."},{"metadata":{"trusted":true},"cell_type":"code","source":"def make_train_gen(dataframe, target_size_dim, x_col, y_col,batch_size=64, my_train_idg=my_image_augmentation(train=True)):\n    train_gen = my_train_idg.flow_from_dataframe(dataframe=dataframe,  \n                                                x_col = x_col,\n                                                y_col = y_col,\n                                                class_mode=\"categorical\",\n                                                target_size=(target_size_dim, target_size_dim), \n                                                 color_mode='rgb',\n                                                batch_size = batch_size)\n\n    return train_gen\n\n\ndef make_val_gen(dataframe, target_size_dim, x_col, y_col,batch_size=64, my_val_idg=my_image_augmentation(train=False)):\n    \n    val_gen = my_val_idg.flow_from_dataframe(dataframe = dataframe, \n                                              x_col = x_col,\n                                              y_col = y_col,\n                                              class_mode=\"categorical\",\n                                              target_size=(target_size_dim, target_size_dim), \n                                              batch_size = batch_size,\n                                                shuffle=False) \n    \n    return val_gen\n\ndef make_test_gen(dataframe, target_size_dim, x_col,batch_size=64, my_test_idg=my_image_augmentation(train=False)):\n    \n    test_gen = my_test_idg.flow_from_dataframe(dataframe=dataframe,\n                                                x_col=x_col,\n                                                y_col=None,\n                                                batch_size=batch_size,\n                                                seed=SEED,\n                                                shuffle=False,\n                                                class_mode=None,\n                                                target_size=(target_size_dim, target_size_dim))\n    return test_gen","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Some important parameters for our configuration"},{"metadata":{},"cell_type":"markdown","source":"The input size comes from Keras blog which recommends an input size of 300 for EfficientNetB3\n\n![Screenshot%202020-12-05%20at%2011.34.08%20PM.png](attachment:Screenshot%202020-12-05%20at%2011.34.08%20PM.png)","attachments":{"Screenshot%202020-12-05%20at%2011.34.08%20PM.png":{"image/png":"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"}}},{"metadata":{"trusted":true},"cell_type":"code","source":"target_size_dim = 300\nbatch_size = 32","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Train Valid Split"},{"metadata":{"trusted":true},"cell_type":"code","source":"df['label'] = df['label'].astype('str') # Since we are using inbuilt generator it takes label as string\n\nX_train, X_valid = train_test_split(df, test_size = 0.1, random_state=SEED, shuffle=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_gen = make_train_gen(X_train, x_col = 'path', y_col='label', batch_size=batch_size, target_size_dim=target_size_dim)\nvalid_gen = make_val_gen(X_valid, x_col = 'path', y_col='label', batch_size=batch_size*2, target_size_dim=target_size_dim)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Visualizing Output of generator"},{"metadata":{"trusted":true},"cell_type":"code","source":"t_x, t_y = next(train_gen)\nfig, m_axs = plt.subplots(4, 6, figsize = (32, 16))\nfor (c_x, c_y, c_ax) in zip(t_x, t_y, m_axs.flatten()):\n    c_ax.imshow(c_x.astype(np.uint8))\n    c_ax.set_title(np.argmax(c_y))\n    c_ax.axis('off')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"t_x, t_y = next(valid_gen)\nfig, m_axs = plt.subplots(4, 6, figsize = (32, 16))\nfor (c_x,  c_ax) in zip(t_x, m_axs.flatten()):\n    c_ax.imshow(c_x.astype(np.uint8))\n    c_ax.set_title(np.argmax(c_y))\n    c_ax.axis('off')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Creating Test Generator"},{"metadata":{"trusted":true},"cell_type":"code","source":"test_gen = make_test_gen(df_test, target_size_dim = target_size_dim, x_col='path', batch_size=batch_size*2)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Creating Model"},{"metadata":{"trusted":true},"cell_type":"code","source":"## Only available in tf2.3+\n\nfrom tensorflow.keras.applications import EfficientNetB3 \nfrom tensorflow.keras.losses import CategoricalCrossentropy","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def load_pretrained_model(weights_path, drop_connect, target_size_dim, layers_to_unfreeze=5):\n    model = EfficientNetB3(\n            weights=None, \n            include_top=False, \n            input_shape=(target_size_dim, target_size_dim, 3),\n            drop_connect_rate=0.4\n        )\n    \n    model.load_weights(weights_path)\n    \n    model.trainable = True\n\n    # for layer in model.layers[-layers_to_unfreeze:]:\n    #     if not isinstance(layer, tf.keras.layers.BatchNormalization): \n    #         layer.trainable = True\n\n    if DEBUG:\n        for layer in model.layers:\n            print(layer.name, layer.trainable)\n\n    return model\n\ndef build_my_model(base_model, optimizer, loss='categorical_crossentropy', metrics = ['categorical_accuracy']):\n    \n    my_model = Sequential()    \n    my_model.add(base_model)\n    my_model.add(GlobalAveragePooling2D())\n    my_model.add(Dense(256))\n    my_model.add(BatchNormalization())\n    my_model.add(Activation('relu'))\n    my_model.add(Dropout(0.3))\n    my_model.add(Dense(5, activation='softmax'))\n    my_model.compile(\n        optimizer=optimizer,\n        loss=CategoricalCrossentropy(label_smoothing=0.05),\n        metrics=metrics\n    )\n    return my_model\n\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#!wget https://storage.googleapis.com/keras-applications/efficientnetb3_notop.h5\n## to get model weights","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model_weights_path = '../input/noisystudent/efficientnetb3_notop.h5'\nmodel_weights_path","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"drop_rate = 0.4 ## value of dropout to be used in loaded network\nbase_model = load_pretrained_model( model_weights_path, drop_rate, target_size_dim )\n\noptimizer = tf.keras.optimizers.Adam(lr = 1e-4)\nmy_model = build_my_model(base_model, optimizer)\nmy_model.summary()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Callbacks"},{"metadata":{"trusted":true},"cell_type":"code","source":"weight_path_save = 'best_model.hdf5'\nlast_weight_path = 'last_model.hdf5'\n\ncheckpoint = ModelCheckpoint(weight_path_save, \n                             monitor= 'val_loss', \n                             verbose=1, \n                             save_best_only=True, \n                             mode= 'min', \n                             save_weights_only = False)\ncheckpoint_last = ModelCheckpoint(last_weight_path, \n                             monitor= 'val_loss', \n                             verbose=1, \n                             save_best_only=False, \n                             mode= 'min', \n                             save_weights_only = False)\n\n\nearly = EarlyStopping(monitor= 'val_loss', \n                      mode= 'min', \n                      patience=10)\n\nreduceLROnPlat = ReduceLROnPlateau(monitor='val_loss', factor=0.8, patience=2, verbose=1, mode='auto', epsilon=0.0001, cooldown=5, min_lr=0.00001)\ncallbacks_list = [checkpoint, checkpoint_last, early, reduceLROnPlat]","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Train Model"},{"metadata":{"trusted":true},"cell_type":"code","source":"if DEBUG:\n    epochs = 3\nelse:\n    epochs = 20","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# from sklearn.utils import class_weight\n\n# classes_to_predict =[0, 1, 2, 3, 4]\n# class_weights = class_weight.compute_class_weight(\"balanced\", classes_to_predict, train_gen.labels)\n# class_weights_dict = {i : class_weights[i] for i,label in enumerate(classes_to_predict)}\n\n# print(class_weights_dict)\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"if DEBUG:\n    history = my_model.fit(train_gen, \n                          validation_data = valid_gen, \n                          epochs = epochs, \n                          callbacks = callbacks_list,\n                           steps_per_epoch=1\n                           #class_weight=class_weights_dict\n                          )\nelse:\n    history = my_model.fit(train_gen, \n                          validation_data = valid_gen, \n                          epochs = epochs, \n                          callbacks = callbacks_list,\n                           #class_weight=class_weights_dict\n                          )","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def plot_hist(hist):\n    plt.figure(figsize=(15,5))\n    plt.plot(np.arange(epochs), hist.history[\"categorical_accuracy\"], '-o', label='Train Accuracy',color='#ff7f0e')\n    plt.plot(np.arange(epochs), hist.history[\"val_categorical_accuracy\"], '-o',label='Val Accuracy',color='#1f77b4')\n    plt.xlabel('Epoch',size=14)\n    plt.ylabel('Accuracy',size=14)\n    plt.legend(loc=2)\n    \n    plt2 = plt.gca().twinx()\n    plt2.plot(np.arange(epochs) ,history.history['loss'],'-o',label='Train Loss',color='#2ca02c')\n    plt2.plot(np.arange(epochs) ,history.history['val_loss'],'-o',label='Val Loss',color='#d62728')\n    plt.legend(loc=3)\n    plt.ylabel('Loss',size=14)\n    plt.title(\"Model Accuracy and loss\")\n    \n    #plt.legend([\"train\", \"validation\"], loc=\"upper left\")\n    \n    plt.savefig('loss.png')\n    plt.show()\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_hist(history)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Evaluating Model on Validation Set"},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.metrics import confusion_matrix, classification_report","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"my_model.load_weights(weight_path_save) ## load the best model or all your metrics would be on the last run not on the best one","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pred_valid_y = my_model.predict(valid_gen,  verbose = True)\npred_valid_y_labels = np.argmax(pred_valid_y, axis=-1)\nvalid_labels=valid_gen.labels\n\nprint(classification_report(valid_labels, pred_valid_y_labels ))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(confusion_matrix(valid_labels, pred_valid_y_labels ))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Getting Predictions on Test Set"},{"metadata":{"trusted":true},"cell_type":"code","source":"pred_test = my_model.predict(test_gen, verbose = True)\npred_test_labels = np.argmax(pred_test, axis = -1)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Creating Submission File"},{"metadata":{"trusted":true},"cell_type":"code","source":"final_submission = df_test\n\nfinal_submission['image_id'] = final_submission.path.str.split('/').str[-1]\nfinal_submission['label'] = pred_test_labels\n\nfinal_csv = final_submission[['image_id', 'label']]\nfinal_csv.head()\n\n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"final_csv.to_csv('submission.csv', index=False)\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"**Work In Progress. I am thinking on why the validation loss is fluctuating. This indicates me to the fact that the dataset is too noisy. Will surely work on some other approach for this noisy dataset. If you learnt something from this kernel kindly upvote :)**"},{"metadata":{},"cell_type":"markdown","source":"## Inference kernel is [here](https://www.kaggle.com/harveenchadha/efficientnetb3-baseline-inference-keras-tf2)"}],"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}