{"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":"2022-06-27T00:41:32.955368Z","iopub.execute_input":"2022-06-27T00:41:32.955784Z","iopub.status.idle":"2022-06-27T00:41:32.966929Z","shell.execute_reply.started":"2022-06-27T00:41:32.955749Z","shell.execute_reply":"2022-06-27T00:41:32.96576Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import pandas as pd\n\nimport tensorflow as tf\nfrom tensorflow import keras\n\nfrom sklearn.model_selection import train_test_split\n\nfrom keras.models import Sequential\nfrom tensorflow.keras import layers\nfrom keras.layers import Dense, Dropout, Flatten, BatchNormalization\nfrom keras.callbacks import ReduceLROnPlateau\nfrom keras.preprocessing.image import ImageDataGenerator\n\nimport warnings \nwarnings.filterwarnings('ignore')","metadata":{"execution":{"iopub.status.busy":"2022-06-27T03:41:34.787637Z","iopub.execute_input":"2022-06-27T03:41:34.788721Z","iopub.status.idle":"2022-06-27T03:41:34.796224Z","shell.execute_reply.started":"2022-06-27T03:41:34.788669Z","shell.execute_reply":"2022-06-27T03:41:34.795192Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train = pd.read_csv(\"../input/digit-recognizer/train.csv\")\ntest = pd.read_csv(\"../input/digit-recognizer/test.csv\")\nsample = pd.read_csv('../input/digit-recognizer/sample_submission.csv')","metadata":{"execution":{"iopub.status.busy":"2022-06-27T00:41:34.618613Z","iopub.execute_input":"2022-06-27T00:41:34.619274Z","iopub.status.idle":"2022-06-27T00:41:39.236658Z","shell.execute_reply.started":"2022-06-27T00:41:34.619238Z","shell.execute_reply":"2022-06-27T00:41:39.235493Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"#This is my first effort, using a simple, densely connected feed forward model. It attained a score of 0.97014. \n\n# Y_train = train[\"label\"]\n# X_train = train.drop(labels = [\"label\"],axis = 1) \n\n# X_train = X_train / 255.0\n# test = test / 255.0\n\n# x_train, y_train, x_val, y_val = train_test_split(X_train, Y_train, test_size = 0.15)\n\n# epochs = 50\n# learning_rate = 0.01\n# decay_rate = learning_rate / epochs\n# momentum = 0.8\n\n# model = keras.models.Sequential()\n# model.add(keras.layers.Dense(300, activation = 'LeakyReLU'))\n# model.add(BatchNormalization())\n# model.add(Dropout(0.5))\n# model.add(keras.layers.Dense(200, activation = 'LeakyReLU'))\n# model.add(BatchNormalization())\n# model.add(Dropout(0.5))\n# model.add(keras.layers.Dense(100, activation = 'LeakyReLU'))\n# model.add(keras.layers.Dense(10, activation = 'softmax'))\n\n# opt = keras.optimizers.SGD(learning_rate=learning_rate, momentum=momentum, decay=decay_rate, nesterov=False)\n# model.compile(loss = 'sparse_categorical_crossentropy', optimizer = opt, metrics = ['accuracy'])\n\n# fit = model.fit(x_train, x_val, batch_size = 128, \n#                     epochs = 240, validation_data =(y_train, y_val))\n\n# pred = model.predict(test)\n# pred_class = np.argmax(pred, axis=-1)\n# out = sample.copy()\n# out['Label'] = pred_class\n# out.to_csv(\"submission.csv\", index=False)","metadata":{"execution":{"iopub.status.busy":"2022-06-26T23:53:33.542613Z","iopub.execute_input":"2022-06-26T23:53:33.543942Z","iopub.status.idle":"2022-06-26T23:53:33.555959Z","shell.execute_reply.started":"2022-06-26T23:53:33.543888Z","shell.execute_reply":"2022-06-26T23:53:33.554328Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"Y_train = train[\"label\"]\nX_train = train.drop(labels = [\"label\"],axis = 1) \n\nX_train = X_train / 255.0\ntest = test / 255.0\n\nX_train = X_train.values.reshape(-1,28,28,1)\ntest = test.values.reshape(-1,28,28,1)\n\nx_train, y_train, x_val, y_val = train_test_split(X_train, Y_train, test_size = 0.15)","metadata":{"execution":{"iopub.status.busy":"2022-06-27T00:41:39.238463Z","iopub.execute_input":"2022-06-27T00:41:39.238818Z","iopub.status.idle":"2022-06-27T00:41:39.850689Z","shell.execute_reply.started":"2022-06-27T00:41:39.238787Z","shell.execute_reply":"2022-06-27T00:41:39.849515Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# data_augmentation = keras.Sequential(\n#     [layers.RandomRotation(0.05),\n#      layers.RandomZoom(0.10),\n#      layers.RandomTranslation(height_factor=(-0.1, 0.1),\n#                               width_factor=(-0.1, 0.1),\n#                               fill_mode='nearest')])","metadata":{"execution":{"iopub.status.busy":"2022-06-27T02:48:59.500122Z","iopub.execute_input":"2022-06-27T02:48:59.500534Z","iopub.status.idle":"2022-06-27T02:48:59.51566Z","shell.execute_reply.started":"2022-06-27T02:48:59.5005Z","shell.execute_reply":"2022-06-27T02:48:59.514585Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# learning_rate_reduction = ReduceLROnPlateau(monitor='val_acc', \n#                                             patience=3, \n#                                             verbose=1, \n#                                             factor=0.5, \n#                                             min_lr=0.00001)","metadata":{"execution":{"iopub.status.busy":"2022-06-27T01:07:36.318206Z","iopub.execute_input":"2022-06-27T01:07:36.318562Z","iopub.status.idle":"2022-06-27T01:07:36.324227Z","shell.execute_reply.started":"2022-06-27T01:07:36.318532Z","shell.execute_reply":"2022-06-27T01:07:36.323022Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"epochs = 50\nlearning_rate = 0.01\ndecay_rate = learning_rate / epochs\nmomentum = 0.8\nbatch_size = 64","metadata":{"execution":{"iopub.status.busy":"2022-06-27T03:57:49.924918Z","iopub.execute_input":"2022-06-27T03:57:49.925495Z","iopub.status.idle":"2022-06-27T03:57:49.931529Z","shell.execute_reply.started":"2022-06-27T03:57:49.925453Z","shell.execute_reply":"2022-06-27T03:57:49.930462Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"datagen = ImageDataGenerator(\n        featurewise_center=False,  # set input mean to 0 over the dataset\n        samplewise_center=False,  # set each sample mean to 0\n        featurewise_std_normalization=False,  # divide inputs by std of the dataset\n        samplewise_std_normalization=False,  # divide each input by its std\n        zca_whitening=False,  # apply ZCA whitening\n        rotation_range=10,  # randomly rotate images in the range (degrees, 0 to 180)\n        zoom_range = 0.1, # Randomly zoom image \n        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)\n        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)\n        horizontal_flip=False,  # randomly flip images\n        vertical_flip=False)  # randomly flip images\n\ndatagen.fit(x_train)","metadata":{"execution":{"iopub.status.busy":"2022-06-27T03:41:45.523414Z","iopub.execute_input":"2022-06-27T03:41:45.523791Z","iopub.status.idle":"2022-06-27T03:41:45.646176Z","shell.execute_reply.started":"2022-06-27T03:41:45.523759Z","shell.execute_reply":"2022-06-27T03:41:45.644881Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"inputs = keras.Input(shape=(28,28,1))\n# x = data_augmentation(inputs)\nx = layers.Conv2D(filters=128, kernel_size=5, activation = 'relu')(inputs)\nx = BatchNormalization()(x)\nx = layers.MaxPooling2D(pool_size=2)(x)\nx = layers.Conv2D(filters=256, kernel_size=3, activation = 'relu')(x)\nx = Dropout(rate = 0.25)(x)\nx = BatchNormalization()(x)\nx = layers.MaxPooling2D(pool_size=2)(x)\nx = layers.Conv2D(filters=256, kernel_size=3, activation = 'relu')(x)\nx = Dropout(rate = 0.25)(x)\nx = BatchNormalization()(x)\nx = layers.MaxPooling2D(pool_size=2)(x)\nx = layers.Flatten()(x)\nx = Dense(256, activation = 'relu')(x)\nx = BatchNormalization()(x)\nx = Dropout(rate = 0.5)(x)\n\noutputs = layers.Dense(10, activation='softmax')(x)\nmodel = keras.Model(inputs=inputs, outputs=outputs)\n\nopt = keras.optimizers.RMSprop(learning_rate=learning_rate, momentum=momentum, decay=decay_rate)\nmodel.compile(loss='sparse_categorical_crossentropy', optimizer=opt, metrics=['accuracy'])","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"fit = model.fit_generator(datagen.flow(x_train,x_val, batch_size=64),\n                              epochs = 30, validation_data = (y_train,y_val),\n                              verbose = 2)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# inputs = keras.Input(shape=(28,28,1))\n# # x = data_augmentation(inputs)\n# x = layers.Conv2D(filters=64, kernel_size=3, activation = 'relu')(inputs)\n# x = BatchNormalization()(x)\n# x = layers.MaxPooling2D(pool_size=2)(x)\n# x = layers.Conv2D(filters=128, kernel_size=3, activation = 'relu')(x)\n# x = Dropout(rate = 0.25)(x)\n# x = layers.MaxPooling2D(pool_size=2)(x)\n# x = layers.Conv2D(filters=128, kernel_size=3, activation = 'relu')(x)\n# x = BatchNormalization()(x)\n# x = layers.MaxPooling2D(pool_size=2)(x)\n# x = Dropout(rate = 0.25)(x)\n\n\n# x = layers.Flatten()(x)\n# x = Dense(256, activation = 'relu')(x)\n# x = BatchNormalization()(x)\n# x = Dropout(rate = 0.5)(x)\n\n# outputs = layers.Dense(10, activation='softmax')(x)\n# model = keras.Model(inputs=inputs, outputs=outputs)\n\n# opt = keras.optimizers.SGD(learning_rate=learning_rate, momentum=momentum, decay=decay_rate, nesterov=False)\n# # opt = keras.optimizers.RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.0)\n# model.compile(loss='sparse_categorical_crossentropy', optimizer=opt, metrics=['accuracy'])","metadata":{"execution":{"iopub.status.busy":"2022-06-27T04:03:56.938249Z","iopub.execute_input":"2022-06-27T04:03:56.938612Z","iopub.status.idle":"2022-06-27T04:03:57.056925Z","shell.execute_reply.started":"2022-06-27T04:03:56.938582Z","shell.execute_reply":"2022-06-27T04:03:57.055734Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# fit = model.fit_generator(datagen.flow(x_train,x_val, batch_size=batch_size),\n#                               epochs = 43, validation_data = (y_train,y_val),\n#                               verbose = 2)","metadata":{"execution":{"iopub.status.busy":"2022-06-27T04:04:00.732874Z","iopub.execute_input":"2022-06-27T04:04:00.734083Z","iopub.status.idle":"2022-06-27T04:50:13.38264Z","shell.execute_reply.started":"2022-06-27T04:04:00.734005Z","shell.execute_reply":"2022-06-27T04:50:13.381557Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"execution":{"iopub.status.busy":"2022-06-27T03:45:15.626396Z","iopub.execute_input":"2022-06-27T03:45:15.626777Z","iopub.status.idle":"2022-06-27T03:45:15.633621Z","shell.execute_reply.started":"2022-06-27T03:45:15.626746Z","shell.execute_reply":"2022-06-27T03:45:15.632798Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"pred = model.predict(test)\npred_class = np.argmax(pred, axis=-1)\nout = sample.copy()\nout['Label'] = pred_class\nout.to_csv(\"submission.csv\", index=False)","metadata":{"execution":{"iopub.status.busy":"2022-06-27T00:14:18.459201Z","iopub.execute_input":"2022-06-27T00:14:18.459595Z","iopub.status.idle":"2022-06-27T00:14:39.363021Z","shell.execute_reply.started":"2022-06-27T00:14:18.45956Z","shell.execute_reply":"2022-06-27T00:14:39.361785Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}