I get pretty bad results on the val set when I run the evaluation as mentioned in the README using the checkpoint 34 provided. I have changed the batch size from 28 to 7. I have attached the screenshot from the output file here:
Can you tell me what can be the issue? I am not really sure, thanks .
Hi, sorry for the late reply, could you change the batch size of each GPU to 1? And do you use spconv2x? Our code is based on spconv1x. You can refer to issue 2 if you want to use spconv2x.
Hi,
I get pretty bad results on the val set when I run the evaluation as mentioned in the README using the checkpoint 34 provided. I have changed the batch size from 28 to 7. I have attached the screenshot from the output file here:
Can you tell me what can be the issue? I am not really sure, thanks .