Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support exporting > 2Gb transformer models #1514

Merged
merged 14 commits into from
May 11, 2023
Merged

Support exporting > 2Gb transformer models #1514

merged 14 commits into from
May 11, 2023

Conversation

dbogunowicz
Copy link
Contributor

@dbogunowicz dbogunowicz commented Apr 10, 2023 •
edited
Loading

Since are now supporting big ONNX models, we are:

  • enforcing that when exporting such a model, its weights will be dumped to model.data file
  • setting tokenizer.json to an optional file

The rationale for making tokenizer.json optional was explained by @natuan in the thread a while ago.

@dbogunowicz dbogunowicz changed the base branch from main to feature/damian/adapt_large_models April 26, 2023 20:17
@dbogunowicz dbogunowicz marked this pull request as ready for review April 26, 2023 20:24
@dbogunowicz dbogunowicz changed the title [LLM Export] Support exporting > 2Gb models Support exporting > 2Gb transformer models Apr 26, 2023
bfineran
bfineran previously approved these changes May 5, 2023
View reviewed changes
Base automatically changed from feature/damian/adapt_large_models to main May 11, 2023 05:43
@dbogunowicz dbogunowicz merged commit b96a89a into main May 11, 2023
12 checks passed
@dbogunowicz dbogunowicz deleted the feature/damian/large_model_export branch May 11, 2023 12:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@bfineran bfineran bfineran approved these changes

@KSGulin KSGulin KSGulin approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
3 participants
@dbogunowicz @bfineran @KSGulin