TypeError：__init__() 为参数“轴”获取了多个值

Question

I want to use attention mechanism with the code bellow:

attention = Dot([decoder_outputs, encoder_outputs], axes=[2, 1])
attention = Activation('softmax')(attention)

context = Dot([attention, encoder_outputs], axes=[2,1])
decoder_combined_context = concatenate([context, decoder_outputs])

But i have this error message and i don't know how to fix it

Traceback (most recent call last):

    attention = Dot([decoder_outputs, encoder_outputs], axes=[2, 1])

TypeError: __init__() got multiple values for argument 'axes'

I'm using keras version 2.3.1 with tensorflow version 2.1.0

Answer 1

The first argument to Dot is called axes :

tf.keras.layers.Dot(axes, normalize=False, **kwargs)

So you need to decide what value you want to pass for axes : either [decoder_outputs, encoder_outputs] or [2, 1] .

Answer 2

You are not calling the layer correctly, it should be like this:

attention = Dot(axes=[2, 1])([decoder_outputs, encoder_outputs])

Your second call to Dot has the same issue, it follows the same pattern