I installed the ONNEX-tensorRT backend GitHub - onnx/onnx-tensorrt: ONNX-TensorRT: TensorRT backend for ONNX in the tensorRT Docker 19.02-py3, generated the trt engine file (yolov3.trt) with the yolov3_onnx sample:
python onnx_to_tensorrt.py
Then I copied the yolov3.trt file into the Deepstream Docker(nvcr.io/nvidia/deepstream:3.0-18.11), changed the config file:
/root/DeepStream_Release/samples/configs/deepstream-app/source4_720p_dec_infer-resnet_tracker_sgie_tiled_display_int8.txt
and
/root/DeepStream_Release/samples/configs/deepstream-app/config_infer_primary.txt
to use my yolov3.trt engine plan file.
when I run the sample script :
deepstream-app -c ./configs/deepstream-app/source4_720p_dec_infer-resnet_tracker_sgie_tiled_display_int8_bak.txt
got the following error:
The engine plan file is incompatible with this version of TensorRT, expecting 5.0.2.3got 5.0.2.6, please rebuild.
My question is:
How can I update TensorRT Version in the DeepStream SDK?
or
Can I update TensorRT Version in the tensorRT Docker?