NVIDIA / NVIDIA/TensorRT

How to reduce reformat layer in QAT???

Open
#4,039 10 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

Description

I want to finetune a quantized yolo model, and export to TRT.
I carefully read the QDQ document and some existed issues to place and remove unused QDQ nodes, the model have 92% int8 precision layers but still have 70 reformat layers.

Environment

TensorRT Version: 8.5.0.2

NVIDIA GPU: agx orin 16GB

NVIDIA Driver Version:

CUDA Version: 11.4

CUDNN Version:

Operating System: Ubuntu 20.04

Python Version (if applicable): 3.8.10

Tensorflow Version (if applicable):

PyTorch Version (if applicable):

Baremetal or Container (if so, version):

Relevant Files

trex

Model link:

model.zip
Use normal trtexec command to convert onnx to engine.

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 the attached model.zip and reproduce the engine using the normal trtexec command described in the issue. Review the QDQ documentation and the TREX visualization to investigate the reported 70 reformat layers. Done means identifying the cause and documenting a verified way to reduce them.

Written by the indexing model from the issue text.

Assessment

Domain
machine-learning, performance
Issue type
Bug
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.