NVIDIA / NVIDIA/TensorRT

how to choose which layers to quant for faster performace?

Open
#3,763 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

triaged
Dominant language
C++
Stars
13.4k
Forks
2.4k
Avg merge
5d 3h
Merged PRs (30d)
2

Description

in the process of yolov8 int8 quant, i find that some layers(int8) is slower than fp16, and the reformat operation is very time-consuming, for best presion, we can do sensitive-layer analysise to get the proper layer to quant , but for best speed, how should i du to identify which layer to quant?
(some screenshot blow)
image

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 provides no repository file, test, or entry point; it only describes YOLOv8 INT8 quantization, slower INT8 layers, and costly reformat operations. Start by locating TensorRT quantization and layer-format selection documentation or examples, then benchmark mixed INT8/FP16 layer choices. Done would require a documented method for identifying the fastest layer combination.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning, performance
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.