[Documentation] Python tutorial missing PyTorch export guidance and external data file handling

Open Beginner friendly
#28,763 5 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
2/5
Estimated time
1-3 hours
Newbie friendliness
74/100
Issue type
Documentation
Clarity
Mostly clear
Activity status
Active
Tech stack
python, pytorch

Research direction

Start with the export-pytorch-model.html page and the Python inference tutorial linked in the issue. Verify the documented behavior for large PyTorch models with opset 17+ and external model data, then explain the companion .onnx.data requirement and how to merge the split files for a self-contained deployment file.

Written by the indexing model from the issue text.

Description

documentation
Describe the documentation issue

The Python inference tutorial (https://onnxruntime.ai/docs/api/python/tutorial.html) only covers sklearn → ONNX. The PyTorch export tutorial
(https://onnxruntime.ai/docs/tutorials/export-pytorch-model.html) only covers custom operators.

Neither page documents a common production gotcha: exporting large PyTorch models with opset 17+ automatically produces two files (model.onnx + model.onnx.data). InferenceSession fails if only model.onnx is present without the companion .data file.

Also undocumented: how to merge the split files into a single, self-contained .onnx file for distribution/deployment.

Page / URL

https://onnxruntime.ai/docs/tutorials/export-pytorch-model.html

Dominant language
C++
Stars
21.9k
Forks
4.2k
Avg merge
3d 23h
Merged PRs (30d)
172

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from microsoft/onnxruntime

All issues in microsoft/onnxruntime

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.