Model pruning #54
Open
Model pruning #54
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
need to re-edit this and check the latest notes
adds pruning from coremltools into exporter. needs to be implemented into main.py for adding it in parser args. also custom set target sparsity. I imagine would be able to also add palletization here as well.
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Outdated suggestions cannot be applied.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.
Specifically the convert.py file
Made for Optimization: cache past key-values, [https://github.com/huggingface/swift-transformers/issues/9]
Prunes model utilizing this. believe it does work, but I haven't gotten a complete model yet because of memory error.
This is also where you may add in palletization. Also including this into the main.py file utilizing user customization for the sparsity and pruning.