How to convert .etlt model to .plan model to use the TAO trained model in triton inference server

I have a classification model trained using TAO toolkit, I exported the model as tensorRT engine model. The resulting model was resnet_035.etlt

I went through model repository structure for inference server, Where I found that for tensorRT models to be used it needs to have following structure:

Screenshot from 2021-12-27 14-50-36

How to convert this .etlt model to .plan model?

anyone could help on this? I’ve trained the .etlt models, and didn’t see doc for how to integrate it to triton server.

Please refer to GitHub - NVIDIA-AI-IOT/tao-toolkit-triton-apps: Sample app code for deploying TAO Toolkit trained models to Triton
and tao-toolkit-triton-apps/download_and_convert.sh at main · NVIDIA-AI-IOT/tao-toolkit-triton-apps · GitHub

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.