sst2_lstm.py 文件源码

def sst2_run(index_embedding, dataset, num_words=5000, embedding_len=100, max_len=50):

    (x_train, y_train), (x_test, y_test), (x_dev, y_dev) = ds.load_data(dataset, num_words, has_dev=True)
    x_train = sequence.pad_sequences(x_train, maxlen=max_len)
    x_test = sequence.pad_sequences(x_test, maxlen=max_len)
    x_dev = sequence.pad_sequences(x_dev, maxlen=max_len)

    model = Sequential()
    model.add(Embedding(num_words, embedding_len, input_length=max_len, weights=[index_embedding]))
    model.add(LSTM(max_len, dropout=0.5, recurrent_dropout=0.5))
    model.add(Dense(1, activation='sigmoid', 
                    kernel_regularizer=regularizers.l2(0.01),
                    activity_regularizer=regularizers.l1(0.01)
            ))

    model.compile(loss='binary_crossentropy',
                    optimizer='adam',
                    metrics=['accuracy'])

    print(model.summary())
    model.fit(x_train, y_train, epochs=4, batch_size=32,
              validation_data=(x_dev, y_dev),  verbose=2)

    score, acc = model.evaluate(x_test, y_test, verbose=0)
    print('Test score:', score)
    print('Test accuracy:', acc)