crash when creating object detection trt model

[kaishijeng]

I use your sd image and build a OpenAdas from the master branch. When I run ./OpenADAS, I got a crash below:

Loading TensorRT engine file at: models/traffic_sign/traffic_sign_classification_resnet18_64.engine 15 TensorRT plan file not found. Creating a new plan file at: models/object_detection/ctdet_bdd_resnet18_384.engine

Input filename: models/object_detection/ctdet_bdd_resnet18_384.onnx ONNX IR version: 0.0.4 Opset version: 9 Producer name: pytorch Producer version: 1.1 Domain: Model version: 0 Doc string:

WARNING: ONNX model has a newer ir_version (0.0.4) than this parser was built against (0.0.3). setFp16Mode Begin building engine... terminate called after throwing an instance of 'thrust::system::system_error' what(): parallel_for failed: no kernel image is available for execution on the device [1] 12143 abort ./OpenADAS

Any idea why this happens?

Thanks,

[lmirel]

Make sure you have the latest code and model then review this part in the README:

Compile and Run
Update GPU_ARCHS: Modify GPU_ARCHS in CMakeLists.txt to suit your GPU. For Jetson Nano, GPU_ARCHS = 53 and for my RTX 2070, GPU_ARCHS = 75. 

It might help.