fasttext_stacking.py 文件源码

def build_model(cat, loss):
    print('Build model...')
    model = Sequential()

    # we start off with an efficient embedding layer which maps
    # our vocab indices into embedding_dims dimensions
    model.add(Embedding(max_features,
                        embedding_dims,
                        input_length=maxlen))

    model.add(Dropout(0.5))

    # we add a GlobalAveragePooling1D, which will average the embeddings
    # of all words in the document
    model.add(GlobalAveragePooling1D())

    model.add(Dropout(0.5))

    # We project onto a single unit output layer, and squash it with a sigmoid:
    model.add(Dense(cat, activation='softmax'))

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

    return model