Skip to content

Enable large models with generate_artifacts API#20942

Closed
carzh wants to merge 3 commits intomicrosoft:mainfrom
carzh:carzh/optional_shape_inference
Closed

Enable large models with generate_artifacts API#20942
carzh wants to merge 3 commits intomicrosoft:mainfrom
carzh:carzh/optional_shape_inference

Conversation

@carzh
Copy link
Copy Markdown
Contributor

@carzh carzh commented Jun 5, 2024

Description

Added enable_large_models flag to generate_artifacts API call

Motivation and Context

  • ONNX function calls such as shape_inference and the ONNX checker run into errors when passed a ModelProto > 2GB (issue tracked here)
  • This PR adds the option to disable these API calls in generate_artifacts until the ONNX serialization issue is resolved

@carzh carzh requested review from baijumeswani and skottmckay June 5, 2024 18:47
@carzh carzh closed this Jun 6, 2024
@carzh carzh deleted the carzh/optional_shape_inference branch June 6, 2024 23:02
carzh added a commit that referenced this pull request Jun 21, 2024
…facts API (#20958)

### Description
Alternative design from #20942 

Allow users to pass in a model path for the generate_artifacts API. 

### Motivation and Context
- ONNX API calls such as the onnx checker + shape inference fail when
given a model > 2GB, but work if a path to a model >2GB is passed in.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant