Failed to import pytorch_quantization when creating onnx file

[kimcheolhee80]

My development environment is as follows. -Jetson AGX Orin -JetPack 5.1.1 -DeepStream 6.2

I followed the instructions below to use yolo-nas. (https://github.com/marcoslucianops/DeepStream-Yolo/blob/master/docs/YOLONAS.md)

The onnx file was created with the $ python3 export_yolonas.py -m yolo_nas_l -w yolo_nas_l_coco.pt command.

While creating the onnx file, the following error log was output, but the onnx file was created normally. Deepstream-app -c xxx.txt confirmed that the video works normally.

Can I ignore Failed to import pytorch_quantization ? I'm a newbie so the question may be stupid. hope you understand.

Thank you very much for your efforts.

------------------------ log -------------------------

$ python3 export_yolonas.py -m yolo_nas_l -w yolo_nas_l_coco.pt h The console stream is logged into /home/dosong01/sg_logs/console.log [2023-06-28 12:44:38] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it [2023-06-28 12:44:42] WARNING - init.py - Failed to import pytorch_quantization [2023-06-28 12:44:42] WARNING - calibrator.py - Failed to import pytorch_quantization [2023-06-28 12:44:42] WARNING - export.py - Failed to import pytorch_quantization [2023-06-28 12:44:42] WARNING - selective_quantization_utils.py - Failed to import pytorch_quantization

Starting: yolo_nas_l_coco.pth Opening YOLO-NAS model

[2023-06-28 12:44:44] INFO - checkpoint_utils.py - Successfully loaded model weights from yolo_nas_l_coco.pth checkpoint.

Exporting the model to ONNX ================ Diagnostic Run torch.onnx.export version 2.0.1 ================ verbose: False, log level: Level.ERROR ======================= 0 NONE 0 NOTE 0 WARNING 0 ERROR =================== =====

Done: yolo_nas_l_coco.onnx

[marcoslucianops]

There's no problem. You can ignore it.