pytorch-ssd icon indicating copy to clipboard operation
pytorch-ssd copied to clipboard

Published 20 hours ago • verified publisher icon qfgaohao

VGG16-SSD with multiple GPU

Open say2sarwar opened this issue 6 years ago • 2 comments
trafficstars

Thank you for open source project. I was training Mobilenet with ssd on multiple GPUs and it's working with some small modification to the code. But, VGG16-SDD have some issues when run on multiple GPUs. Did you test your code on More than one GPUs?

Thanks in advance.

say2sarwar avatar Nov 18 '19 01:11 say2sarwar

I also want to know how to fix this. I followed some advice and add "net = torch.nn.DataParallel(net)", then I got following error: RuntimeError: Expected tensor for argument #1 'input' to have the same device as tensor for argument #2 'weight'; but device 1 does not equal 0 (while checking arguments for cudnn_batch_norm) No idea what to do.

AiueoABC avatar Apr 20 '20 09:04 AiueoABC

I have the exact same error with vgg16-SSD, did anyone got a clue on how to fix this issue?

Gcardoso233 avatar Sep 22 '21 16:09 Gcardoso233