NVIDIA / NVIDIA/TensorRT

MAP difference between TensorRT 8 and TensorRT10 for YOLOV8 DLA Compilation

Open
#4,697 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

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

Description

Description

Environment

TensorRT Version: 8.5.2.2

CUDA Version: 11.4.315

CUDNN Version: 8.6.0.166

**Nvidia Jetson Orin NX 16GB

Operating System:

Python Version (if applicable): 3.8.10

Used Ultralytics Jetpack5 latest docker for the testing

Relevant Files

Model link:
yolovm8 custom trained model for 3 classes

Have you tried the latest release?:

Attach the captured .json and .bin files from TensorRT's API Capture tool if you're on an x86_64 Unix system

Can this model run on other frameworks? For example run ONNX model with ONNXRuntime (polygraphy run <model.onnx> --onnxrt):

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 by reproducing the YOLOv8 custom-model validation on the stated Orin NX setup with TensorRT 8.5.2.2 and compare it with the TensorRT 10 result. Run the ONNX model with ONNXRuntime as suggested, and collect the model, validation details, and TensorRT API capture files if available. Done means identifying whether the near-zero MAP is caused by the model, validation path, or TensorRT/DLA version difference.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
embedded-iot, machine-learning
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.