Where can I find sample config_infer_*.txt file to deploy .onnx model exported by TAO 5.5.0 in Deepstream 6.4?

Please provide the following information when requesting support.

• Hardware (T4/V100/Xavier/Nano/etc): dGPU
• Network Type (Detectnet_v2/Faster_rcnn/Yolo_v4/LPRnet/Mask_rcnn/Classification/etc): Classification
• TLT Version (Please run “tlt info --verbose” and share “docker_tag” here): TAO 5.5.0
• Training spec file(If have, please share here): NA
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.): NA

I have a.onnx files of models trained using TAO 5.5.0 and I want to deploy it with Deepstream 6.4, I cannot find an example config_infer_*.txt file for .onnx model, all available examples for for deploying .etlt files.

• Example provided on deepstream_tao_apps repo: deepstream_tao_apps/deepstream_app_tao_configs/nvinfer/config_infer_secondary_vehiclemakenet.txt at release/tao_ds7.1ga · NVIDIA-AI-IOT/deepstream_tao_apps · GitHub. But this one is for Deepstream 7.1 not Deepstream 6.4.

On this page it mentions the following:

So I looked for those sample files. Example provided by Deepstream 6.4:

• config_infer_secondary_vehiclemake.txt (3.5 KB)
• config_infer_secondary_vehicletypes.txt (3.5 KB)

Both of them are for .etlt files, not .onnx files.

The Deepstream 6.4 doc for TAO Toolkit Integration with DeepStream mentions using.etlt files, no mention of .onnx files.

Where can I find a sample config_infer_*.txt file to deploy .onnx files for Classification models directly in Deepstream 6.4?

2 posts - 2 participants

Read full topic