Recurrent Attention (Step 2): Support functional composition of RNN Cells (incl. constants) #7981
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
I think we should add support for optional named inputs (states, consts etc) for layers at topology level and should be reflected in the |
| @@ -4,6 +4,8 @@ | |||
| import functools | |||
| import warnings | |||
|
|
|||
| from keras.engine import Model | |||
| from keras.layers.wrappers import Wrapper | |||
There was a problem hiding this comment.
Use relative imports.
|
We will keep all additions related to recurrent attention in a single PR: #7980, closing this one. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR is the second step to support recurrent attention mechanisms as suggested in #7633.
In short, it adds the possibility to compose RNN Cells using keras functional API, including support of constants, by introducing the
Model-like wrapper classFunctionalRNNCell.Note: this is an extension of #7980