{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"df = pd.read_csv('../input/train.csv')\ndf_test = pd.read_csv('../input/test.csv')\ndf.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"140fc6c582127f3fe3616072470f4e129cbc0971"},"cell_type":"code","source":"from sklearn.model_selection import train_test_split\nfrom keras.preprocessing.text import Tokenizer\nfrom keras.preprocessing.sequence import pad_sequences\n\n\nX_train, X_test = train_test_split(df, test_size=0.1, random_state=2018)\n\n# config values\nembed_size = 300 # how big is each word vector\nmax_features = 50000 # how many unique words to use (i.e num rows in embedding vector)\nmaxlen = 100 # max number of words in a question to use\n\ny_train, y_test = X_train['target'].values, X_test['target'].values\n\nX_train = X_train['question_text'].fillna('_NA_').values\nX_test = X_test['question_text'].fillna('_NA_').values\nX_submission = df_test['question_text'].fillna('_NA_').values\n\ntokenizer = Tokenizer(num_words=max_features)\ntokenizer.fit_on_texts(list(X_train))\nX_train = tokenizer.texts_to_sequences(X_train)\nX_test = tokenizer.texts_to_sequences(X_test)\nX_submission = tokenizer.texts_to_sequences(X_submission)\n\nX_train = pad_sequences(X_train, maxlen=maxlen)\nX_test = pad_sequences(X_test, maxlen=maxlen)\nX_submission = pad_sequences(X_submission, maxlen=maxlen)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"scrolled":true,"_uuid":"ee948289e62708d534f1b79610daf93e359a937d"},"cell_type":"code","source":"from keras.layers import Dense, Input, LSTM, Embedding, Dropout, Activation\nfrom keras.layers import Bidirectional, GlobalMaxPool1D\nfrom keras.models import Model\n\n\ninp = Input(shape=(maxlen,))\nlayer = Embedding(max_features, embed_size)(inp)\nlayer = Bidirectional(LSTM(64, return_sequences=True))(layer)\nlayer = GlobalMaxPool1D()(layer)\nlayer = Dense(16, activation=\"relu\")(layer)\nlayer = Dropout(0.1)(layer)\nlayer = Dense(1, activation=\"sigmoid\")(layer)\nmodel = Model(inputs=inp, outputs=layer)\nmodel.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n\nmodel.summary()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"e527808125f1c14c43ab4c2c47d02ace42ae14e1"},"cell_type":"code","source":"model.fit(X_train, y_train, batch_size=512, epochs=2, validation_data=(X_test, y_test))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"9edd2502a1e3f3769b7f4d4afe8686f78a9ff3ff"},"cell_type":"code","source":"from sklearn import metrics\n\npred_test_y = model.predict([X_test], batch_size=1024, verbose=1)\nfor thresh in np.arange(0.1, 0.501, 0.01):\n    thresh = np.round(thresh, 2)\n    print('F1 score at threshold {} is {}'.format(thresh, metrics.f1_score(y_test, (pred_test_y > thresh).astype(int))))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"ed9f1874c910d943436a98f085b2f6eb61343c74"},"cell_type":"code","source":"pred_submission_y = model.predict([X_submission], batch_size=1024, verbose=1)\npred_submission_y = (pred_submission_y > 0.29).astype(int)\n\ndf_submission = pd.DataFrame({'qid': df_test['qid'].values})\ndf_submission['prediction'] = pred_submission_y\ndf_submission.to_csv(\"submission.csv\", index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"e5c3825061dc3977418612ba54f69ff35906c41c"},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}