search

zhijian-liu / torchpack

A neural network training interface based on PyTorch, with a focus on flexibility
https://pypi.org/project/torchpack/
MIT License
61 stars 15 forks source link

Address already used #35

Closed qfwysw closed 6 months ago

qfwysw commented 2 years ago

image image I print the tcp. From the results, setting the port does not seem to have any effect

zhijian-liu commented 2 years ago

The number after : indicates the number of processes that should be launched on that node. I don't think it's possible to manually set the port, except by changing the value at https://github.com/zhijian-liu/torchpack/blob/master/torchpack/launch/launchers/drunner.py#L73.

jiachens commented 1 year ago

any updates on this issue?

zhijian-liu commented 1 year ago

@jiachens, I haven't got a chance to fix this yet. This should be a minor fix. Feel free to submit a PR if you have time.

zhenghw111 commented 1 year ago

Hello! I have encountered the same problem. How can I solve it?

Sranc3 commented 10 months ago

Have you fixed the bug?

qfwysw commented 10 months ago

It's been a long time, and I've lost track of the specific code. My previous solution was to plagiarize the train.py file of mmdetection3d. Hope this helps.

Sranc3 commented 10 months ago

It's been a long time, and I've lost track of the specific code. My previous solution was to plagiarize the train.py file of mmdetection3d. Hope this helps.

Thanks for your kind suggestion!