more efficient zipvmap implementation - #23
Closed
marcelluethi wants to merge 2 commits into
Closed
Conversation
Collaborator
|
Please see #20, in this clean up I implemented zipvmap based on vmap (for any tuple) |
Contributor
Author
|
I just tested your implementation. It completely solves the performance problem and is more general and elegant than the solution I proposed. I will close this PR. |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The current
zipvmapfunction is implemented to work with abitrary tuple types.As a result it could not use jax vmap function directly. This causes a very significant performance penalty.
This PR proposes to have specialised zipvmap calls for zips of 2,3 and 4 tensors, which are much more efficient.
As a fallback for more then 4 dimensions, the generic method is used.
As the specialised code is rather repetitive and long, I moved it out of the tensorops file into a dedicated module.
The interface remains the same and all the tests still pass.