Execution of convert_to_onnx.py from onnx_packnet directory resulted in displaying Killed

Autonomous Machines Jetson & Embedded Systems Jetson Orin Nano
1

Hi all, I tried executing the script convert_to_onnx.py from the directory /usr/src/tensorrt/samples/python/onnx_packnet/

python3.6 convert_to_onnx.py --output model.onnx

But it resulted in displaying the message Killed and it could not succeed

Below is the log output

/home/mistral/.local/lib/python3.6/site-packages/torch/nn/functional.py:2471: UserWarning: floordiv is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the ‘trunc’ function NOT ‘floor’). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode=‘trunc’), or for actual floor division, use torch.div(a, b, rounding_mode=‘floor’).
_verify_batch_size([input.size(0) * input.size(1) // num_groups, num_groups] + list(input.size()[2:]))
/usr/src/tensorrt/samples/python/onnx_packnet/packnet-sfm-0.1.2/packnet_sfm/networks/layers/packnet/layers01.py:146: UserWarning: floordiv is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the ‘trunc’ function NOT ‘floor’). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode=‘trunc’), or for actual floor division, use torch.div(a, b, rounding_mode=‘floor’).
out_h, out_w = h // r, w // r
Killed

Let me know what could be the reason behind it and the solution for it.

Thanks and Regards

Nagaraj Trivedi

3

Hi,

Usually, killed indicates the device might run out of memory.
Please verify this with tegrastats.

If the system memory does reach its maximum and causes the error.
Please try to add some swap memory to see if it helps.

Thanks.

5

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