pytorch / pytorch/executorch

where can i do the quantize or de-quantize operation in demo

Open
#13,628 4 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

module: quantization
Dominant language
Python
Stars
5k
Forks
1.2k
Avg merge
2d 10h
Merged PRs (30d)
581

Description

🚀 The feature, motivation and pitch

Hello:
In this demo https://github.com/pytorch/executorch/blob/main/examples/portable/executor_runner/executor_runner.cpp
I cant find out the quantize and de-quantize operation in the input_tensors and output_tensors. I guess that they are all float32 precision in input_tensors and output tensors; If so, how can i move the quantize and dequantize out of pte files and i do quantize and dequantize by myself;

Thank you very much
Alternatives

No response

Additional context

No response

RFC (Optional)

No response

cc @kimishpatel @jerryzh168

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

Start with examples/portable/executor_runner/executor_runner.cpp and trace how input_tensors and output_tensors are handled relative to the PTE file. Determine whether quantization and dequantization occur inside the exported model or around the runner, then document how a user could perform those operations externally; the issue is done when the tensor precision and supported workflow are clearly explained.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp, pytorch
Domain
machine-learning
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.