microsoft / microsoft/onnxruntime-inference-examples

How to set the quantize range?

Open
#51 0 comments 0 reactions 1 assignee View on GitHub

@yufenglee is already working on this.

Since Nov 3, 2021.

Dominant language
C++
Stars
1.7k
Forks
414
Avg merge
1d 6h
Merged PRs (30d)
14

Description

Hi,
I am wondering why the quantize range of onnxruntime is fixed to [0,255]? How can I set the quantize range to [-128,127] (https://intellabs.github.io/distiller/algo_quantization.html)?

and how can i set per-channel-scale?

according to the document, U8U8 or U8S8 has better performance on x86-64? What is the reason behind that? What if I use S8U8 or S8S8?

Contributor guide

No contributing guide indexed for this repository

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.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.