You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Apr 11, 2021. It is now read-only.
I use the latest version and check the similarity part of the model, i.e.,
qa_model = merge([question_output, answer_output], mode=similarity, output_shape=lambda x: x[:-1],name='similarity')
When I invoke qa_model.summary(), here is the screen output:
The last layer (similarity) should have an output shape of (None, 1) but is (None, 200). I use the default 'cos' mode of Keras, it is (None,1). Also, the program doesn't work. It has nan loss.
The text was updated successfully, but these errors were encountered:
guxd
changed the title
Output shape of similarity merge layer is still incorrect
Output shape of the similarity merge layer is still incorrect
Jun 30, 2016
I fixed this just now. I think the output shape should just always be (None, 1). The thing is, I don't think it made a difference. I think the nan loss could be coming from somewhere else? Not sure. Let me know if this fixed it.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I use the latest version and check the similarity part of the model, i.e.,
![insuranceqa similarity](https://cloud.githubusercontent.com/assets/6091014/16475683/16d8bdd8-3eb3-11e6-9124-27bed6d4cdb6.png)
qa_model = merge([question_output, answer_output], mode=similarity, output_shape=lambda x: x[:-1],name='similarity')
When I invoke qa_model.summary(), here is the screen output:
The last layer (similarity) should have an output shape of (None, 1) but is (None, 200). I use the default 'cos' mode of Keras, it is (None,1). Also, the program doesn't work. It has nan loss.
The text was updated successfully, but these errors were encountered: