Error in getting attention map

[suyashh94]

When I try to get the attention as shown in https://github.com/CyberZHG/keras-bert/blob/master/demo/load_model/load_and_get_attention_map.py, I get the following error :

Output tensors of a Functional model must be the output of a TensorFlow Layer (thus holding past layer metadata). Found: tf.Tensor

Minimal Codes To Reproduce

inputs = model.inputs[:2]
outputs = (model.get_layer('Encoder-1-MultiHeadSelfAttention').attention)
model = keras.models.Model(inputs , outputs)

** Have tried with the function too as shown in the example and that shows similar error too.

[stale]

Is this still relevant? If so, what is blocking it? Is there anything you can do to help move it forward?

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[suyashh94]

I have implemented it using Google's release of tensorflow.

[stale]

Is this still relevant? If so, what is blocking it? Is there anything you can do to help move it forward?

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.