Hi , I am trying to use the demo in Xavier NX with
python3 demo_trt.py ctdet --demo webcam --arch hardnet_85 --load_model /home/littro/CenterNet-HarDNet/models/centernet_hardnet85_coco.pth --input_w 416 --input_h 416
It can convert model to .onnx but during conversion from .onnx to .trt it have some errors. If you see the error carefully it is about some layer [TensorRT] ERROR: Layer: Where_1305's output can not be used as shape tensor.
According to your suggestion I also tried pytorch 1.6.0 and 1.5.0.
I also tried to convert .onnx model to .trt using trtexec in TensorRT. but the error is same about Layer: Where_1305's.
You can see my systems specs libraries and error in the attached screenshots.
Looking forward to hearing from you. Thank you.
Hi , I am trying to use the demo in Xavier NX with
python3 demo_trt.py ctdet --demo webcam --arch hardnet_85 --load_model /home/littro/CenterNet-HarDNet/models/centernet_hardnet85_coco.pth --input_w 416 --input_h 416It can convert model to .onnx but during conversion from .onnx to .trt it have some errors. If you see the error carefully it is about some layer
[TensorRT] ERROR: Layer: Where_1305's output can not be used as shape tensor.According to your suggestion I also tried
pytorch 1.6.0 and 1.5.0.I also tried to convert .onnx model to .trt using
trtexecin TensorRT. but the error is same aboutLayer: Where_1305's.You can see my systems specs libraries and error in the attached screenshots.
Looking forward to hearing from you. Thank you.