{"cells":[{"metadata":{"id":"wYtuKeK0dImp","trusted":true},"cell_type":"code","source":"import csv\nimport numpy as np\nimport tensorflow as tf\nfrom tensorflow.keras.preprocessing.image import ImageDataGenerator","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import pandas as pd\ndf_train = pd.read_csv(\"../input/digit-recognizer/train.csv\")\ndf_test = pd.read_csv(\"../input/digit-recognizer/test.csv\")\ndf_train.head()","execution_count":null,"outputs":[]},{"metadata":{"id":"4kxw-_rmcnVu","trusted":true},"cell_type":"code","source":"train_labels = df_train['label']\ndel df_train['label']\ndf_train.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_set = df_train.values\ntrain_set = train_set.reshape(42000,28,28)\ntest_set = df_test.values\ntest_set = test_set.reshape(28000,28,28)\nprint(train_set.shape, test_set.shape)","execution_count":null,"outputs":[]},{"metadata":{"id":"awoqRpyZdQkD","trusted":true},"cell_type":"code","source":"# Create an ImageDataGenerator and do Image Augmentation\ntrain_datagen = ImageDataGenerator(rescale=1.0/255,\n                                   rotation_range=10,\n                                   width_shift_range=0.1,\n                                   height_shift_range=0.1,\n                                   zoom_range=0.1)\n\nvalidation_datagen = ImageDataGenerator(rescale=1.0/255)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"num = 10\nmodel = [0]*num\nfrom tensorflow.keras import regularizers\n# from tensorflow.keras.callbacks import ReduceLROnPlateau\n# ankitz = ReduceLROnPlateau(monitor='val_accuracy', patience = 3, verbose=1,factor=0.5, min_lr=0.00001)\n\nfor i in range(num):\n    model[i] = tf.keras.models.Sequential([\n        tf.keras.layers.Conv2D(32, (5, 5), activation='relu', input_shape=(28, 28, 1),padding='same'),\n        tf.keras.layers.Conv2D(32, (5, 5), activation='relu', padding='same'),\n        tf.keras.layers.MaxPool2D(pool_size=(2, 2),strides=(2,2)),\n        tf.keras.layers.Dropout(0.2),\n        tf.keras.layers.Conv2D(64, (5, 5), activation='relu', padding='same'),\n        tf.keras.layers.Conv2D(64, (5, 5), activation='relu', padding='same'),\n        tf.keras.layers.MaxPool2D(pool_size=(2, 2),strides=(2,2)),\n        tf.keras.layers.Dropout(0.2),\n        tf.keras.layers.Flatten(),\n        tf.keras.layers.Dense(256, activation=tf.nn.relu),\n        tf.keras.layers.Dropout(0.2),\n        tf.keras.layers.Dense(10, activation='softmax')])\n    model[i].compile(optimizer=tf.optimizers.Adam(), loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n\nmodel[0].summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from tensorflow.keras.callbacks import ReduceLROnPlateau\nfrom sklearn.model_selection import train_test_split\n\nepochs = 25\nbatch_size = 86\nhistory = [0]*num\nfor i in range(num):\n    random_seed = 7\n    training_set, validation_set, training_labels, validation_labels = train_test_split(train_set, train_labels, test_size = 0.10, random_state=random_seed)\n    training_images = np.expand_dims(training_set, axis=-1)\n    testing_images = np.expand_dims(test_set, axis=-1)\n    validation_images = np.expand_dims(validation_set, axis=-1)\n    ankitz = ReduceLROnPlateau(monitor='val_accuracy', patience = 2, verbose=1,factor=0.7, min_lr=0.00001)\n    history[i] = model[i].fit_generator(train_datagen.flow(training_images, training_labels, batch_size=batch_size),\n                              epochs = epochs,\n                              steps_per_epoch=training_images.shape[0] // batch_size,\n                              validation_data=validation_datagen.flow(validation_images, validation_labels),\n                             callbacks = [ankitz])\n    print(\"Model \"+str(i+1)+\" has finished training..!!\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def predict(X_data):\n    results = np.zeros((X_data.shape[0],10))\n    for j in range(num):\n        results = results + model[j].predict(X_data)\n    return results\nresults = predict(testing_images)\nresults = np.argmax(results,axis = 1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"results = pd.Series(results,name=\"Label\")\nsubmission = pd.concat([pd.Series(range(1,28001),name = \"ImageId\"),results],axis = 1)\nprint(submission[:40])\nsubmission.to_csv(\"my_results_ensemble.csv\",index=False)","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}