{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"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 in \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 \"../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# Any results you write to the current directory are saved as output.","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"data = pd.read_csv('/kaggle/input/Kannada-MNIST/train.csv')\n# .iloc[:,1:] drops labels and retains just pixel data\ny, x = data['label'].values, data.iloc[:,1:].values.astype(float)/255\n# Reshaping for tf batch optimization\nx = x.reshape(-1,28,28,1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test = pd.read_csv('/kaggle/input/Kannada-MNIST/test.csv')\nx_test = test.iloc[:,1:].values.astype(float)/255","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"dig = pd.read_csv('../input/Kannada-MNIST/Dig-MNIST.csv')\nx_dig = dig.iloc[:,1:].values.astype(float)/255\nx_dig = x_dig.reshape(-1,28,28,1)\ny_dig = dig.label.values","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"x_test = x_test.reshape(-1,28,28,1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.preprocessing import OneHotEncoder","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Transform target into onehot\nohe = OneHotEncoder(handle_unknown='ignore')\ny = ohe.fit_transform(y[:,np.newaxis])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import tensorflow as tf\n#Make sure to utilize the gpu though this shouldn't be necessary\ntf.device('/GPU:0')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Build model. Construction is on the heavy side but because competition only takes one forward pass it shouldn't matter. Doesn't close in on the time limit\nmodel = tf.keras.Sequential([tf.keras.layers.Conv2D(input_shape=(28,28,1),activation ='relu', filters = 621, kernel_size = 3, padding='same'),\n                             tf.keras.layers.Dropout(0.4),\n                             tf.keras.layers.Conv2D(488,3,padding = 'same'),\n                             tf.keras.layers.Dropout(0.4),\n                             tf.keras.layers.Conv2D(388,3,padding = 'same'),\n                             tf.keras.layers.Dropout(0.4),\n                             tf.keras.layers.Conv2D(276,3,padding = 'same'),\n                             tf.keras.layers.Dropout(0.4),\n                             tf.keras.layers.Flatten(),\n                             tf.keras.layers.Dense(300,activation = 'relu'),\n                             tf.keras.layers.Dropout(0.4),\n                             tf.keras.layers.Dense(10, activation = 'softmax')\n    \n])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Compile the model. Adam for stochastic gradient descent.\nmodel.compile(optimizer = 'adam',\n             loss = 'categorical_crossentropy',\n             metrics = ['acc'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Fit model. Usually takes only couple epochs to find solution. Epochs are overdone to allow early stopping to do its magic.\nmodel.fit(x, y,\n          epochs = 100,\n          validation_split = 0.2,\n          callbacks = [tf.keras.callbacks.EarlyStopping(monitor = 'val_acc', mode = 'max', patience = 3, restore_best_weights = True)])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Test model against the \"actual\" test set \"dig\"\ndig_pred = tf.keras.backend.eval(tf.argmax(model.predict(x_dig),1))\nnp.sum(dig_pred==y_dig)/y_dig.shape[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Make competition predictions and write the result to csv\npred = tf.keras.backend.eval(tf.argmax(model.predict(x_test),1))\nout = pd.DataFrame({'id': test['id'], 'label': pred})\nout.to_csv(\"submission.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":1}