Unable to train custom data

Robotics & Edge Computing Jetson & Embedded Systems Jetson Orin Nano
1

I am facing issues when training it my custom images to proceed.

.onnx file created successfully but while I am using the command detectnet --model=models/marker/ssd-mobilenet.onnx --labels=models/marker/labels.txt --input-blob=input_0 --output-cvg=scores --output-bbox=boxes /dev/video0 , it crashes by showing the message “Segmentation fault (core dumped).

Please help me out to solve this issue.

3

Hi,

Could you share which software release you use?
Due to the API changes in TensorRT 10, please setup JetPack 6.0 or prior release to run the jetson-inference sample.

Thanks.

4

Hello,

Thanks for your response.

Installed Lib:
Jetpack: 6.2
CUDA: 12.8.61
cuDNN: 9.3.0.75
TensorRT: 10.3.0.30
OpenCV: 4.8.0 with CUDA: No

5

Hi,

Most of the models inside the jetson-inference are caffe-based.
So it’s recommended to set up the device with JetPack 6.0 as our latest software drops Caffe model support.

If you want to use JetPack 6.2, please try it with the ONNX model.
You will also need some modifications since there are some API changes in the TensorRT 10.

You can find some info in the link below:

Thanks.