Best practices for getting help with TensorRT:
The best way to get help with TensorRT is by opening an issue on the GitHub:
- For any new feature request, bugs or issues, please go to GitHub - NVIDIA/TensorRT: NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT., select “issues” tap and click on “New Issue”
When filing the issue, please make sure you:
- Please mention New Feature Request in the description if you request new feature/support in TRT
- If you don’t know if this is new feature, please put details about your issues in the description
- Provide details on the platforms you’re using:
- Linux distro and version
- GPU type
- Nvidia driver version
- CUDA version
- CUDNN version
- Python version [if using python]
- Tensorflow version
- TensorRT version
- If Jetson, also include OS & hw versions
- Describe the issue
- Provide supporting code, model or upload data files for reproducing the issues
Refer to Github OSS contribution rules for more details: TensorRT/CONTRIBUTING.md at release/10.9 · NVIDIA/TensorRT · GitHub