microsoft / microsoft/onnxruntime

Capability to export tensors in C++

Open
#10,679 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature request
Dominant language
C++
Stars
21.9k
Forks
4.2k
Avg merge
4d 8h
Merged PRs (30d)
179

Description

I'm trying to use a model created in pytorch using onnxruntime, we are seeing off behaviors from the model when comparing it to how it performs in python. To fully understand what is going on, it would be useful to allow exporting input and output tensors so they can be tested out in a different platform too.

Is there currently a way to export input/output tensors to a .pt file? - Similar to torch.save() in python

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.

Research direction

The issue names no files, tests, or entry points. Start by reviewing the existing C++ tensor APIs and how input and output tensors are represented, then compare the requested .pt export with Python's torch.save(). Done should mean a defined way to export both tensor types for testing on another platform.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.