improvement of PositionEmbeddingPosition #2192
Closed
Conversation
This file contains hidden or 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
|
@mattdangerw May you take a review? |
|
@divyashreepathihalli It seems to have run out of time. |
|
@divyashreepathihalli |
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Outdated suggestions cannot be applied.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.
PositionEmbedding is widely used in many models in keras_hub.
But he has a problem, which is that it will limit the input length. And in actual use, it seems to be a very troublesome thing to want to change this length. I hope it can be improved.
Specifically, I hope to add a calculation method. When the input length is greater than the preset sequence_length, the calculation method is as follows:
With proper handling, this is lossless. The part less than sequence_length length is exactly the same as the original