SRFBN_CVPR19 icon indicating copy to clipboard operation
SRFBN_CVPR19 copied to clipboard

Published 20 hours ago • verified publisher icon Paper99

some problems?

Open Peterisfar opened this issue 5 years ago • 1 comments

1、why make the png to npy? 2、when i make the json‘s gpu_id=[0,1],it only support the batchsize=2?when i make the batchsize=4, out of memory?but the gpu only use 1G/10G？i don't know why? 3、when i add the os.environ["CUDA_VISIBLE_DEVICES"] = "0, 1" in train.py , batchsize can be 4, it can be work! but i make batchsize is 16. the gpu only use 2G/10G, i don't konw why? the image is 480*270.

Peterisfar avatar Jun 13 '19 12:06 Peterisfar

1: For fast data I/O. 2, 3: We will test the compatibility of multi-gpu and give your feedback in a few days.

Paper99 avatar Jun 25 '19 05:06 Paper99