Removed the unnecessary softplus in NTMHeadBase._address_memory #6
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
Removed the softplus in the softmax:
```python
s = F.softmax(F.softplus(s), dim=1)
```
softmax already constrains the values to (0, +inf), the softplus doesn't achieve much.
|
Makes sense, thanks for that. Can you please run the copy-task notebook and see we're getting the same results? |
|
I trained a bunch of pretty long models, and get good results in the notebooks. |
|
Like this one, which was trained for a while on sequences up to 120 long, and converges very sharply
|
|
Tested it as well, seems to alter convergence a bit but perhaps for the better. |
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Outdated suggestions cannot be applied.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.
Removed the softplus in the softmax:
softmax already constrains the values to (0, 1), the softplus doesn't achieve anything. Pytorch's softmax implementation is already numerically stable, so that's not the preoccupation.