{"cells":[{"metadata":{"_uuid":"57212004253bf93eaef63a728d8839668a618006"},"cell_type":"markdown","source":"## Reference\n* https://www.kaggle.com/suicaokhoailang/magic-numbers-is-all-you-need-0-692-lb Magic numbers is all you need [0.692 LB]\n* https://www.kaggle.com/gmhost/gru-capsule GRU+Capsule\n* Blending with Linear Regression: https://www.kaggle.com/suicaokhoailang/blending-with-linear-regression-0-688-lb\n* Beating the baseline with ONE WEIRD TRICK!: https://www.kaggle.com/suicaokhoailang/beating-the-baseline-with-one-weird-trick-0-691\n* https://www.kaggle.com/shujian/single-rnn-with-4-folds-clr by shujian\n* https://www.kaggle.com/sudalairajkumar/a-look-at-different-embeddings by SRK\n* https://www.kaggle.com/christofhenkel/how-to-preprocessing-when-using-embeddings by Dieter\n* https://www.kaggle.com/shujian/mix-of-nn-models-based-on-meta-embedding by shujian\n* https://www.kaggle.com/gmhost/gru-capsule by Puck Wang\n* Based on SRK's kernel: https://www.kaggle.com/sudalairajkumar/a-look-at-different-embeddings\n* Vladimir Demidov's 2DCNN textClassifier: https://www.kaggle.com/yekenot/2dcnn-textclassifier\n* Attention layer from Khoi Ngyuen: https://www.kaggle.com/suicaokhoailang/lstm-attention-baseline-0-652-lb\n* LSTM model from Strideradu: https://www.kaggle.com/strideradu/word2vec-and-gensim-go-go-go\n* https://www.kaggle.com/danofer/different-embeddings-with-attention-fork"},{"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 \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 the files in the input directory\n\nimport os\nprint(os.listdir(\"../input\"))\n\n# Any results you write to the current directory are saved as output.","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"e31d6e126881ee56a1de3efe02fcf309e900ef00"},"cell_type":"code","source":"## some config values \nembed_size = 300 # how big is each word vector\nmax_features = 95000 # how many unique words to use (i.e num rows in embedding vector)\nmaxlen = 70 # max number of words in a question to use","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"522d9790478f62193ea5c315372a2ab9cbe9b27f"},"cell_type":"markdown","source":"**Load packages and data**"},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"import os\nimport time\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nfrom tqdm import tqdm\nimport math\nfrom sklearn.model_selection import train_test_split\nfrom sklearn import metrics\n\nfrom keras.preprocessing.text import Tokenizer\nfrom keras.preprocessing.sequence import pad_sequences\nfrom keras.layers import Dense, Input, CuDNNLSTM, Embedding, Dropout, Activation, CuDNNGRU, Conv1D\nfrom keras.layers import Bidirectional, GlobalMaxPool1D, GlobalMaxPooling1D, GlobalAveragePooling1D\nfrom keras.layers import Input, Embedding, Dense, Conv2D, MaxPool2D, concatenate\nfrom keras.layers import Reshape, Flatten, Concatenate, Dropout, SpatialDropout1D\nfrom keras.optimizers import Adam\nfrom keras.models import Model\nfrom keras import backend as K\nfrom keras.engine.topology import Layer\nfrom keras import initializers, regularizers, constraints, optimizers, layers","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"e18ed4339c33036cc39ded79c58901da9fbe0aeb"},"cell_type":"code","source":"# puncts = [ '|', \"'\", '$', '&', '/', '[', ']', '>', '%', '=', '#', '*', '+', '\\\\', '•',  '~', '@', '£', \n#  '·', '_', '{', '}', '©', '^', '®', '`',  '<', '→', '°', '€', '™', '›',  '♥', '←', '×', '§', '″', '′', 'Â', '█', '½', 'à',\n#  '“', '★', '”', '–', '●', 'â', '►', '−', '¢', '²', '¬', '░', '¶', '↑', '±', '¿', '▾', '═', '¦', '║', '―', '¥', '▓', '—', '‹', '─', \n#  '▒', '：', '¼', '⊕', '▼', '▪', '†', '■', '’', '▀', '¨', '▄', '♫', '☆', 'é', '¯', '♦', '¤', '▲', 'è', '¸', '¾', 'Ã', '⋅', '‘', '∞', \n#  '∙', '）', '↓', '、', '│', '（', '»', '，', '♪', '╩', '╚', '³', '・', '╦', '╣', '╔', '╗', '▬', '❤', 'ï', 'Ø', '¹', '≤', '‡', '√', \n# 'é', '&amp;','₹', 'á', '²', 'ế', '청', '하', '¨', '‘', '√', '\\xa0', '高', '端', '大', '气', '上', '档', '次', '_', '½', 'π', '#', \n# '小', '鹿', '乱', '撞', '成', '语', 'ë', 'à', 'ç', '@', 'ü', 'č', 'ć', 'ž', 'đ', '°', 'द', 'े', 'श', '्', 'र', 'ो', 'ह', \n# 'ि', 'प', 'स', 'थ', 'त', 'न', 'व', 'ा', 'ल', 'ं', '林', '彪', '€', '\\u200b', '˚', 'ö', '~', '—', '越', '人', 'च', 'म', 'क', \n# 'ु', 'य', 'ी', 'ê', 'ă', 'ễ', '∞', '抗', '日', '神', '剧', '，', '\\uf02d', '–', 'ご', 'め', 'な', 'さ', 'い', 'す', \n# 'み', 'ま', 'せ', 'ん', 'ó', 'è', '£', '¡', 'ś', '≤', '¿', 'λ', '魔', '法', '师', '）', 'ğ', 'ñ', 'ř', '그', '자', '식', '멀', \n# '쩡', '다', '인', '공', '호', '흡', '데', '혀', '밀', '어', '넣', '는', '거', '보', '니', 'ǒ', 'ú', '️', 'ش', 'ه', 'ا', 'د',\n# 'ة', 'ل', 'ت', 'َ', 'ع', 'م', 'ّ', 'ق', 'ِ', 'ف', 'ي', 'ب', 'ح', 'ْ', 'ث', '³', '饭', '可', '以', '吃', '话', '不', '讲', \n# '∈', 'ℝ', '爾', '汝', '文', '言', '∀', '禮', 'इ', 'ब', 'छ', 'ड', '़', 'ʒ', '有', '「', '寧', '錯', '殺', '一', '千', '絕', \n# '放', '過', '」', '之', '勢', '㏒', '㏑', 'ू', 'â', 'ω', 'ą', 'ō', '精', '杯', 'í', '生', '懸', '命', 'ਨ', 'ਾ', 'ਮ', 'ੁ', \n# '₁', '₂', 'ϵ', 'ä', 'к', 'ɾ', '\\ufeff', 'ã', '©', '\\x9d', 'ū', '™', '＝', 'ù', 'ɪ', 'ŋ', 'خ', 'ر', 'س', 'ن', 'ḵ', 'ā']\npuncts = [',', '.', '\"', ':', ')', '(', '-', '!', '?', '|', ';', \"'\", '$', '&', '/', '[', ']', '>', '%', '=', '#', '*', '+', '\\\\', '•',  '~', '@', '£', \n '·', '_', '{', '}', '©', '^', '®', '`',  '<', '→', '°', '€', '™', '›',  '♥', '←', '×', '§', '″', '′', 'Â', '█', '½', 'à', '…', \n '“', '★', '”', '–', '●', 'â', '►', '−', '¢', '²', '¬', '░', '¶', '↑', '±', '¿', '▾', '═', '¦', '║', '―', '¥', '▓', '—', '‹', '─', \n '▒', '：', '¼', '⊕', '▼', '▪', '†', '■', '’', '▀', '¨', '▄', '♫', '☆', 'é', '¯', '♦', '¤', '▲', 'è', '¸', '¾', 'Ã', '⋅', '‘', '∞', \n '∙', '）', '↓', '、', '│', '（', '»', '，', '♪', '╩', '╚', '³', '・', '╦', '╣', '╔', '╗', '▬', '❤', 'ï', 'Ø', '¹', '≤', '‡', '√', ]\ndef clean_text(x):\n    x = str(x)\n    for punct in puncts:\n        x = x.replace(punct, f' {punct} ')\n    return x","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"5cdc95950037613c690c49b27930ae0f59eb23c3"},"cell_type":"code","source":"def load_and_prec():\n    train_df = pd.read_csv(\"../input/train.csv\")\n    test_df = pd.read_csv(\"../input/test.csv\")\n    \n    train_df[\"question_text\"] = train_df[\"question_text\"].str.lower()\n    test_df[\"question_text\"] = test_df[\"question_text\"].str.lower()\n    \n    train_df[\"question_text\"] = train_df[\"question_text\"].apply(lambda x: clean_text(x))\n    test_df[\"question_text\"] = test_df[\"question_text\"].apply(lambda x: clean_text(x))\n    \n    print(\"Train shape : \",train_df.shape)\n    print(\"Test shape : \",test_df.shape)\n    \n    ## split to train and val\n    train_df, val_df = train_test_split(train_df, test_size=0.001, random_state=2018) # hahaha\n\n\n    ## fill up the missing values\n    train_X = train_df[\"question_text\"].fillna(\"_##_\").values\n    val_X = val_df[\"question_text\"].fillna(\"_##_\").values\n    test_X = test_df[\"question_text\"].fillna(\"_##_\").values\n\n    ## Tokenize the sentences\n    tokenizer = Tokenizer(num_words=max_features)\n    tokenizer.fit_on_texts(list(train_X))\n    train_X = tokenizer.texts_to_sequences(train_X)\n    val_X = tokenizer.texts_to_sequences(val_X)\n    test_X = tokenizer.texts_to_sequences(test_X)\n\n    ## Pad the sentences \n    train_X = pad_sequences(train_X, maxlen=maxlen)\n    val_X = pad_sequences(val_X, maxlen=maxlen)\n    test_X = pad_sequences(test_X, maxlen=maxlen)\n\n    ## Get the target values\n    train_y = train_df['target'].values\n    val_y = val_df['target'].values  \n    \n    #shuffling the data\n    np.random.seed(2018)\n    trn_idx = np.random.permutation(len(train_X))\n    val_idx = np.random.permutation(len(val_X))\n\n    train_X = train_X[trn_idx]\n    val_X = val_X[val_idx]\n    train_y = train_y[trn_idx]\n    val_y = val_y[val_idx]    \n    \n    return train_X, val_X, test_X, train_y, val_y, tokenizer.word_index","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"dba1893c267a1e7536bbf720636647d85c7e349c"},"cell_type":"markdown","source":"**Load embeddings**"},{"metadata":{"trusted":true,"_uuid":"a662716cc5fbbcc0c84019a87c52332ed8912e8d"},"cell_type":"code","source":"def load_glove(word_index):\n    EMBEDDING_FILE = '../input/embeddings/glove.840B.300d/glove.840B.300d.txt'\n    def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')\n    embeddings_index = dict(get_coefs(*o.split(\" \")) for o in open(EMBEDDING_FILE))\n\n    all_embs = np.stack(embeddings_index.values())\n    emb_mean,emb_std = -0.005838499,0.48782197\n    embed_size = all_embs.shape[1]\n\n    # word_index = tokenizer.word_index\n    nb_words = min(max_features, len(word_index))\n    embedding_matrix = np.random.normal(emb_mean, emb_std, (nb_words, embed_size))\n    for word, i in word_index.items():\n        if i >= max_features: continue\n        embedding_vector = embeddings_index.get(word)\n        if embedding_vector is not None: embedding_matrix[i] = embedding_vector\n            \n    return embedding_matrix \n    \ndef load_fasttext(word_index):    \n    EMBEDDING_FILE = '../input/embeddings/wiki-news-300d-1M/wiki-news-300d-1M.vec'\n    def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')\n    embeddings_index = dict(get_coefs(*o.split(\" \")) for o in open(EMBEDDING_FILE) if len(o)>100)\n\n    all_embs = np.stack(embeddings_index.values())\n    emb_mean,emb_std = all_embs.mean(), all_embs.std()\n    embed_size = all_embs.shape[1]\n\n    # word_index = tokenizer.word_index\n    nb_words = min(max_features, len(word_index))\n    embedding_matrix = np.random.normal(emb_mean, emb_std, (nb_words, embed_size))\n    for word, i in word_index.items():\n        if i >= max_features: continue\n        embedding_vector = embeddings_index.get(word)\n        if embedding_vector is not None: embedding_matrix[i] = embedding_vector\n\n    return embedding_matrix\n\ndef load_para(word_index):\n    EMBEDDING_FILE = '../input/embeddings/paragram_300_sl999/paragram_300_sl999.txt'\n    def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')\n    embeddings_index = dict(get_coefs(*o.split(\" \")) for o in open(EMBEDDING_FILE, encoding=\"utf8\", errors='ignore') if len(o)>100)\n\n    all_embs = np.stack(embeddings_index.values())\n    emb_mean,emb_std = -0.0053247833,0.49346462\n    embed_size = all_embs.shape[1]\n    print(emb_mean,emb_std,\"para\")\n\n    # word_index = tokenizer.word_index\n    nb_words = min(max_features, len(word_index))\n    embedding_matrix = np.random.normal(emb_mean, emb_std, (nb_words, embed_size))\n    for word, i in word_index.items():\n        if i >= max_features: continue\n        embedding_vector = embeddings_index.get(word)\n        if embedding_vector is not None: embedding_matrix[i] = embedding_vector\n    \n    return embedding_matrix","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"73d68544af4c48bf9ee37492ecd05feb0b494351"},"cell_type":"markdown","source":"**CNN Model**"},{"metadata":{"trusted":true,"_uuid":"1f72c8c9573fb840ceb50a9cd4ac4e455e1c0ea7"},"cell_type":"code","source":"# https://www.kaggle.com/yekenot/2dcnn-textclassifier\ndef model_cnn(embedding_matrix):\n    filter_sizes = [1,2,3,5]\n    num_filters = 36\n\n    inp = Input(shape=(maxlen,))\n    x = Embedding(max_features, embed_size, weights=[embedding_matrix])(inp)\n    x = Reshape((maxlen, embed_size, 1))(x)\n\n    maxpool_pool = []\n    for i in range(len(filter_sizes)):\n        conv = Conv2D(num_filters, kernel_size=(filter_sizes[i], embed_size),\n                                     kernel_initializer='he_normal', activation='elu')(x)\n        maxpool_pool.append(MaxPool2D(pool_size=(maxlen - filter_sizes[i] + 1, 1))(conv))\n\n    z = Concatenate(axis=1)(maxpool_pool)   \n    z = Flatten()(z)\n    z = Dropout(0.1)(z)\n\n    outp = Dense(1, activation=\"sigmoid\")(z)\n\n    model = Model(inputs=inp, outputs=outp)\n    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n    \n    return model","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"5a676c3a275514a3351edf306e02d832a5f39317"},"cell_type":"markdown","source":"**Attention layer**"},{"metadata":{"trusted":true,"_uuid":"84e00df2c7b94205f5588af503f62412c48f46f3"},"cell_type":"code","source":"# https://www.kaggle.com/suicaokhoailang/lstm-attention-baseline-0-652-lb\n\nclass Attention(Layer):\n    def __init__(self, step_dim,\n                 W_regularizer=None, b_regularizer=None,\n                 W_constraint=None, b_constraint=None,\n                 bias=True, **kwargs):\n        self.supports_masking = True\n        self.init = initializers.get('glorot_uniform')\n\n        self.W_regularizer = regularizers.get(W_regularizer)\n        self.b_regularizer = regularizers.get(b_regularizer)\n\n        self.W_constraint = constraints.get(W_constraint)\n        self.b_constraint = constraints.get(b_constraint)\n\n        self.bias = bias\n        self.step_dim = step_dim\n        self.features_dim = 0\n        super(Attention, self).__init__(**kwargs)\n\n    def build(self, input_shape):\n        assert len(input_shape) == 3\n\n        self.W = self.add_weight((input_shape[-1],),\n                                 initializer=self.init,\n                                 name='{}_W'.format(self.name),\n                                 regularizer=self.W_regularizer,\n                                 constraint=self.W_constraint)\n        self.features_dim = input_shape[-1]\n\n        if self.bias:\n            self.b = self.add_weight((input_shape[1],),\n                                     initializer='zero',\n                                     name='{}_b'.format(self.name),\n                                     regularizer=self.b_regularizer,\n                                     constraint=self.b_constraint)\n        else:\n            self.b = None\n\n        self.built = True\n\n    def compute_mask(self, input, input_mask=None):\n        return None\n\n    def call(self, x, mask=None):\n        features_dim = self.features_dim\n        step_dim = self.step_dim\n\n        eij = K.reshape(K.dot(K.reshape(x, (-1, features_dim)),\n                        K.reshape(self.W, (features_dim, 1))), (-1, step_dim))\n\n        if self.bias:\n            eij += self.b\n\n        eij = K.tanh(eij)\n\n        a = K.exp(eij)\n\n        if mask is not None:\n            a *= K.cast(mask, K.floatx())\n\n        a /= K.cast(K.sum(a, axis=1, keepdims=True) + K.epsilon(), K.floatx())\n\n        a = K.expand_dims(a)\n        weighted_input = x * a\n        return K.sum(weighted_input, axis=1)\n\n    def compute_output_shape(self, input_shape):\n        return input_shape[0],  self.features_dim","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d96793d88c22274d985436e192f62970c227c324"},"cell_type":"markdown","source":"**LSTM models**"},{"metadata":{"trusted":true,"_uuid":"05164d541a0c35cae727d0338548d156efe21427"},"cell_type":"code","source":"def model_lstm_atten(embedding_matrix):\n    inp = Input(shape=(maxlen,))\n    x = Embedding(max_features, embed_size, weights=[embedding_matrix], trainable=False)(inp)\n    x = Bidirectional(CuDNNLSTM(128, return_sequences=True))(x)\n    x = Bidirectional(CuDNNLSTM(64, return_sequences=True))(x)\n    x = Attention(maxlen)(x)\n    x = Dense(64, activation=\"relu\")(x)\n    x = Dense(1, activation=\"sigmoid\")(x)\n    model = Model(inputs=inp, outputs=x)\n    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n    \n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"2bf953e2b5b6d9363eda89971ecc0ac416e2ddd0"},"cell_type":"code","source":"def model_gru_srk_atten(embedding_matrix):\n    inp = Input(shape=(maxlen,))\n    x = Embedding(max_features, embed_size, weights=[embedding_matrix])(inp)\n    x = Bidirectional(CuDNNGRU(64, return_sequences=True))(x)\n    x = Attention(maxlen)(x) # New\n    x = Dense(16, activation=\"relu\")(x)\n    x = Dropout(0.1)(x)\n    x = Dense(1, activation=\"sigmoid\")(x)\n    model = Model(inputs=inp, outputs=x)\n    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n    \n    return model    \n    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"17b13ef39fbbf1919307c23efd516eddc2135023"},"cell_type":"code","source":"def model_lstm_du(embedding_matrix):\n    inp = Input(shape=(maxlen,))\n    x = Embedding(max_features, embed_size, weights=[embedding_matrix])(inp)\n    x = Bidirectional(CuDNNGRU(64, return_sequences=True))(x)\n    avg_pool = GlobalAveragePooling1D()(x)\n    max_pool = GlobalMaxPooling1D()(x)\n    conc = concatenate([avg_pool, max_pool])\n    conc = Dense(64, activation=\"relu\")(conc)\n    conc = Dropout(0.1)(conc)\n    outp = Dense(1, activation=\"sigmoid\")(conc)\n    \n    model = Model(inputs=inp, outputs=outp)\n    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"c02ace553c962a10f038d21801c8b46f567a0c3f"},"cell_type":"code","source":"def model_gru_atten_3(embedding_matrix):\n    inp = Input(shape=(maxlen,))\n    x = Embedding(max_features, embed_size, weights=[embedding_matrix], trainable=False)(inp)\n    x = Bidirectional(CuDNNGRU(128, return_sequences=True))(x)\n    x = Bidirectional(CuDNNGRU(100, return_sequences=True))(x)\n    x = Bidirectional(CuDNNGRU(64, return_sequences=True))(x)\n    x = Attention(maxlen)(x)\n    x = Dense(1, activation=\"sigmoid\")(x)\n    model = Model(inputs=inp, outputs=x)\n    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n    \n    return model","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"a8c857424e9c9f1703a71c1c0ade28713314dd29"},"cell_type":"markdown","source":"**Train and predict**"},{"metadata":{"trusted":true,"_uuid":"e8523d876b6eae762e673b777cc7af4d7f085792"},"cell_type":"code","source":"# https://www.kaggle.com/strideradu/word2vec-and-gensim-go-go-go\ndef train_pred(model, epochs=2):\n    for e in range(epochs):\n        model.fit(train_X, train_y, batch_size=512, epochs=1, validation_data=(val_X, val_y))\n        pred_val_y = model.predict([val_X], batch_size=1024, verbose=0)\n    pred_test_y = model.predict([test_X], batch_size=1024, verbose=0)\n    return pred_val_y, pred_test_y","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"f79081928ca032fbfe3b90c6d3ce91cf57d443d8"},"cell_type":"markdown","source":"**Main part: load, train, pred and blend**"},{"metadata":{"trusted":true,"_uuid":"99d03d2eb63600f1b222522616eab3fa35819f37"},"cell_type":"code","source":"train_X, val_X, test_X, train_y, val_y, word_index = load_and_prec()\nvocab = []\nfor w,k in word_index.items():\n    vocab.append(w)\n    if k >= max_features:\n        break\nembedding_matrix_1 = load_glove(word_index)\n# embedding_matrix_2 = load_fasttext(word_index)\nembedding_matrix_3 = load_para(word_index)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"f0ea9b1468bd7cd3ceead2593c641900dc3a2a77"},"cell_type":"code","source":"## Simple average: http://aclweb.org/anthology/N18-2031\n\n# We have presented an argument for averaging as\n# a valid meta-embedding technique, and found experimental\n# performance to be close to, or in some cases \n# better than that of concatenation, with the\n# additional benefit of reduced dimensionality  \n\n\n## Unweighted DME in https://arxiv.org/pdf/1804.07983.pdf\n\n# “The downside of concatenating embeddings and \n#  giving that as input to an RNN encoder, however,\n#  is that the network then quickly becomes inefficient\n#  as we combine more and more embeddings.”\n  \n# embedding_matrix = np.mean([embedding_matrix_1, embedding_matrix_2, embedding_matrix_3], axis = 0)\nembedding_matrix = np.mean([embedding_matrix_1, embedding_matrix_3], axis = 0)\nnp.shape(embedding_matrix)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"da641bf6a7e0a40863e571c78eba9285fd8671b0"},"cell_type":"code","source":"outputs = []","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"8b0ce7a68d32a893affb0f3e57e0d1517776e1b1"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_gru_atten_3(embedding_matrix), epochs = 3)\noutputs.append([pred_val_y, pred_test_y, '3 GRU w/ atten'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"7aae47c7ded0b4f4849fe68b8c2f282ea03e1d20"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_gru_srk_atten(embedding_matrix), epochs = 2)\noutputs.append([pred_val_y, pred_test_y, 'gru atten srk'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"40ec64ad93aea2eb4d303fb7c1e9fc58590885dc"},"cell_type":"code","source":"# pred_val_y, pred_test_y = train_pred(model_cnn(embedding_matrix), epochs = 2)\n# outputs.append([pred_val_y, pred_test_y, '2d CNN'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"2d5802cdc286219cbddd4caa9851d08cdf39ebf6"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_cnn(embedding_matrix_1), epochs = 2) # GloVe only\noutputs.append([pred_val_y, pred_test_y, '2d CNN GloVe'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"2ce89a82185e091178a3878fb87deddba8e7a381"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_lstm_du(embedding_matrix), epochs = 2)\noutputs.append([pred_val_y, pred_test_y, 'LSTM DU'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"57af85779fca0bfebc26813de4f4d07137e68510"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_lstm_atten(embedding_matrix), epochs = 3)\noutputs.append([pred_val_y, pred_test_y, '2 LSTM w/ attention'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"d7865acc1dfd4074ad9736f0563c7a53eab2bdfc"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_lstm_atten(embedding_matrix_1), epochs = 3) # Only GloVe\noutputs.append([pred_val_y, pred_test_y, '2 LSTM w/ attention GloVe'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"6d0d468740e54fa4ae52262a54179e58b64ef7da"},"cell_type":"code","source":"pred_val_y, pred_test_y = train_pred(model_lstm_atten(embedding_matrix_3), epochs = 3) # Only Para\noutputs.append([pred_val_y, pred_test_y, '2 LSTM w/ attention Para'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"74a84952d3a6a2b265dcd72a1ba8f41e315df332"},"cell_type":"code","source":"# pred_test_y = np.sum([outputs[i][1] * weights[i] for i in range(len(outputs))], axis = 0)\ncoefs = [0.20076554,0.07993707,0.11611663,0.14885248,0.15734404,0.17454667,0.14288361]\n# pred_test_y = np.mean([outputs[i][1] for i in range(len(outputs))], axis = 0)\npred_test_y = np.sum([outputs[i][1]*coefs[i] for i in range(len(coefs))], axis = 0)\n\npred_test_y = (pred_test_y > 0.34).astype(int)\ntest_df = pd.read_csv(\"../input/test.csv\", usecols=[\"qid\"])\nout_df = pd.DataFrame({\"qid\":test_df[\"qid\"].values})\nout_df['prediction'] = pred_test_y\nout_df.to_csv(\"submission.csv\", index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"af44b57ac7afcc8552dd0de65dcecb26538cfc2b"},"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}