alternative solution for keras.layers.Merge

[DevMahmoud10]

In your CFModel.py, you have used keras.layers.Merge() function to apply a dot product to the two embedding layers output and this is deprecated long time ago.
Now, I'm using keras 2.2.0 and tensorflow 1.9.0 as backend and found out that using keras.layers.dot( [X,Y] ,axes=1) is a replacement.

[kyolxs]

In your CFModel.py, you have used keras.layers.Merge() function to apply a dot product to the two embedding layers output and this is deprecated long time ago.
Now, I'm using keras 2.2.0 and tensorflow 1.9.0 as backend and found out that using keras.layers.dot( [X,Y] ,axes=1) is a replacement.

I don't understand the modification of the keras.layers.Merge() function. I tried to modify it, but something went wrong. Can you provide a modified CFModel.py at your convenience?

[ivukasinovic]

Because I can't solve that problem and can't find libraries versions to work well can you write your libraries versions in this project?

[kyolxs]

thx～ Jai Kumar Dewani <notifications@github.com> 于 2020年7月7日周二 19:46写道：

…

instead of from keras.layers import Embedding, Reshape, Merge use from keras.layers import Embedding, Reshape, Concatenate and change the last line while building the model. Instead of self.add(Merge([P, Q], mode='dot', dot_axes=1)) use self.add(Concatenate([P, Q], mode='dot', dot_axes=1)) This should solve your problem — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEUB6L74XEUUVA6V66FZNETR2MDH7ANCNFSM4FKTPJSA> .

[HripsimeS]

If you modify CFModel.py with the code attached, it will work !