{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"# This is my first notebook for Deep Learning. Feedback will be appreciated.","execution_count":null,"outputs":[]},{"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":{"trusted":true},"cell_type":"code","source":"np.random.seed(42)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data_full = pd.read_csv('/kaggle/input/digit-recognizer/train.csv')\nlabels = train_data_full.label\ntrain_data_full.drop(['label'], axis = 1, inplace = True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data_full.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_data_full.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"labels.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"labels.value_counts(sort = False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.model_selection import train_test_split\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.preprocessing import OneHotEncoder\n\nsc = StandardScaler()\nonehot = OneHotEncoder()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_val, y_train, y_val = train_test_split(train_data_full, labels, test_size = 0.2, random_state = 42)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(X_train.shape)\nprint(y_train.shape)\nprint(X_val.shape)\nprint(y_val.shape)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Scaling training and validation sets\nX_train = sc.fit_transform(X_train)\nX_val = sc.transform(X_val)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Encoding the output layer of training set\ny_train = np.array(y_train).reshape(33600, 1)\ny_train = onehot.fit_transform(y_train)\n\n# Encoding the output layer of test set\ny_val = np.array(y_val).reshape(8400, 1)\ny_val = onehot.transform(y_val)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from keras.models import Sequential\nfrom keras.layers import Dense\nfrom keras.layers import Softmax\nfrom keras.layers import Dropout\nfrom keras.optimizers import sgd, RMSprop","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"optimizer = RMSprop(lr=0.0001)\n\nmodel = Sequential()\nmodel.add(Dense(64, input_dim = 784, activation = 'relu', kernel_initializer = 'glorot_normal', bias_initializer = 'zeros'))\nmodel.add(Dropout(0.2))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.0))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.2))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.0))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.0))\nmodel.add(Dense(10, activation = 'softmax'))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.compile(optimizer = optimizer, loss = 'categorical_crossentropy', metrics = ['accuracy'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.fit(X_train, y_train, epochs = 70, batch_size = 64, validation_data = (X_val, y_val))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Taking all the data to train the model\nX_train = train_data_full.copy()\ny_train = labels.copy()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Scaling and Encoding all data\nsc = StandardScaler()\nonehot = OneHotEncoder()\n\nX_train = sc.fit_transform(X_train)\n\ny_train = np.array(y_train).reshape(42000, 1)\ny_train = onehot.fit_transform(y_train)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Training the model again\noptimizer = RMSprop(lr=0.0001)\n\nmodel = Sequential()\nmodel.add(Dense(64, input_dim = 784, activation = 'relu', kernel_initializer = 'glorot_normal', bias_initializer = 'zeros'))\nmodel.add(Dropout(0.2))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.0))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.2))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.0))\nmodel.add(Dense(64, activation = 'relu'))\nmodel.add(Dropout(0.0))\nmodel.add(Dense(10, activation = 'softmax'))\n\nmodel.compile(optimizer = optimizer, loss = 'categorical_crossentropy', metrics = ['accuracy'])\n\nmodel.fit(X_train, y_train, epochs = 70, batch_size = 64)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Loading the test set\ntest_data = pd.read_csv('/kaggle/input/digit-recognizer/test.csv')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_test = sc.transform(test_data)\nX_test.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"predictions = model.predict(X_test)\npredictions.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"predicted_labels = np.argmax(predictions, axis = 1)\nresults = pd.Series(predicted_labels, name = 'Label')\nresults_df = pd.DataFrame()\nresults_df['ImageId'] = [i for i in range(1, 28001)]\nresults_df['Label'] = results","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"results_df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"results_df.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":4}