Keras w/ Tensorflow 批量提取中间层

Question

我目前正在尝试利用我已经训练好的 DL 模型中的中间层作为对给定输入的嵌入。下面的代码已经可以获取我想要的层，但是对于大量输入迭代地执行此操作非常慢。

model = load_model('model.h5')
inp = model.input
outputs = [layer.output for layer in model.layers]
functors = [K.function([inp]+ [K.learning_phase()], [out]) for out in outputs]

def text2tensor(text):
    """Convert string to tensor"""
    tensor = tokenizer.texts_to_sequences([text])
    tensor = pad_sequences(tensor, maxlen=10, padding='pre')
    return tensor

def get_embedding(tensor, at_layer):
    """Get output at particular layer in network """
    functors = [K.function([inp]+ [K.learning_phase()], [out]) for out in outputs][at_layer-1]
    layer_outs = [func([tensor, 1.]) for func in [functors]]
    return layer_outs[0][0]


texts = ['this is my first text',
         'this is my second text',
         'this is my third text',
         .....nth text]

embeddings = np.empty((0,256))
for t in texts:
    tensor = text2tensor(t)
    embedding = get_embedding(tensor,at_layer=4)
    embeddings = np.append(embeddings,[embedding[0]],axis=0)

我如何利用批处理，而不必一个接一个地做这件事？上面的实现速度非常慢，但它有效。

Answer 1

除了我在评论中提到的一点，我建议你创建一个模型而不是后端函数：

input_tensor = Input(shape=(10,))   # assuming maxlen=10
new_model = Model(input_tensor, my_desired_layer.output)

然后，首先预处理您的文本数据以形成输入数组（即下面的 my_data），然后使用 predict 方法并向其传递 batch_size 参数以利用批处理：

out = new_model.predict(my_data)   # the default batch size is 32

Keras w/ Tensorflow 批量提取中间层

Keras w/ Tensorflow intermediate layer extraction in batches

python

batch-processing

deep-learning

keras

tensorflow