Reduce GPU memory for Whisper models converted to ONNX #17378
Merged
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
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
|
Use hash table to speed up like in https://github.com/huggingface/optimum/blob/7450ca30e295abc9e20d56d0aa741402322def0f/optimum/onnx/transformations_utils.py#L31-L54? In that implementation, they ignore those with dimension 1 with data type int32 or int64, or scalar with dimension 0. That could filter out small initializers. Might also help speed up a little. Regarding to |
tianleiwu
previously approved these changes
Sep 1, 2023
tianleiwu
approved these changes
Sep 5, 2023
tianleiwu
pushed a commit
that referenced
this pull request
Oct 31, 2023
### Description This PR changes the Whisper export scripts to further optimize the process of removing duplicate initializers from two subgraphs. The current Greedy approach is quicker by a large factor, but results in some duplicate initializers not being caught and removed. This not only results in a slightly larger Whisper model, but also a model that uses more GPU memory. The approach in this PR uses data hashes and caches to keep a quick export but no longer rely on a greedy approach. --------- Co-authored-by: Peter McAughan <[email protected]>
kleiti
pushed a commit
to kleiti/onnxruntime
that referenced
this pull request
Mar 22, 2024
### Description This PR changes the Whisper export scripts to further optimize the process of removing duplicate initializers from two subgraphs. The current Greedy approach is quicker by a large factor, but results in some duplicate initializers not being caught and removed. This not only results in a slightly larger Whisper model, but also a model that uses more GPU memory. The approach in this PR uses data hashes and caches to keep a quick export but no longer rely on a greedy approach. --------- Co-authored-by: Peter McAughan <[email protected]>
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.
Description
This PR changes the Whisper export scripts to further optimize the process of removing duplicate initializers from two subgraphs.
The current Greedy approach is quicker by a large factor, but results in some duplicate initializers not being caught and removed. This not only results in a slightly larger Whisper model, but also a model that uses more GPU memory.
The approach in this PR uses data hashes and caches to keep a quick export but no longer rely on a greedy approach.