Trtexec use for in orin nano

kalecikli1478 · July 26, 2023, 9:58am

I get trt output with trtexec, but when I try to run this output in v8, I get the error “unicodedecodeerror: ‘utf-8’ codec can’t decode byte 0xe8 in position 4: invalid continuation byte”?
note:
When I convert as yolo…mode=engine, trt happens, but the performance increase I want is not achieved.
Can anyone help with these issues?

AastaLLL · July 27, 2023, 2:25am

Hi,

The error is related to encoding/decoding.
Could you share the API/command that causes this error?

Thanks.

kalecikli1478 · July 27, 2023, 5:15am

/usr/src/tensorrt/bin/trtexec --onnx=v8.onnx --saveEngine=v8.engine
ım converting onnx to engine file with this code. Then when ı try the engine file in my code , ım getting the this error.

AastaLLL · August 3, 2023, 3:28am

Hi,

It seems that you meet the same issue as below.
Please try the workaround mentioned in that topic:

github.com/ultralytics/ultralytics

YOLO v8 model conversion in tensorrt, UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe8 in position 4: invalid continuation byte

opened 05:47PM - 01 Mar 23 UTC

andreazuna89

question non-reproducible Stale

### Search before asking - [X] I have searched the YOLOv8 [issues](https://gith…ub.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question Hi! I am trying to use a yolov8 model converted using tensorrt. When I do the following command: ``` from ultralytics import YOLO results= YOLO('./yolov8s.engine').predict(device=0) ``` I get the following error: ``` Ultralytics YOLOv8.0.44 🚀 Python-3.8.10 torch-2.0.0a0+ec3941ad.nv23.02 CUDA:0 (Orin, 30589MiB) Loading yolov8s.engine for TensorRT inference... Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/root/ultralytics/ultralytics/yolo/engine/model.py", line 201, in predict self.predictor.setup_model(model=self.model) File "/root/ultralytics/ultralytics/yolo/engine/predictor.py", line 221, in setup_model self.model = AutoBackend(model, device=device, dnn=self.args.dnn, data=self.args.data, fp16=self.args.half) File "/root/ultralytics/ultralytics/nn/autobackend.py", line 151, in __init__ meta = json.loads(f.read(meta_len).decode('utf-8')) # read metadata UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe8 in position 4: invalid continuation byte ``` Can someone help me here? Thanks a lot Andrea ### Additional _No response_

Thanks.

system · August 23, 2023, 3:34am

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.