examples icon indicating copy to clipboard operation
examples copied to clipboard

Just get a low accuracy of 75.8 with resnet50 on ImageNet

Open mountain111 opened this issue 1 year ago • 2 comments

I train resnet50 on ImageNet with GPUs=8, batchsize=256, learning-rate=0.1, epochs=90, and momentum=0.90. The attained top1 accuracy is 75.80, lower than the reported 76.15. The gap is not marginal on the large-scale ImageNet. Why does the difference exist?

mountain111 avatar May 19 '23 22:05 mountain111

I get 75.644 with resnet50 on ImageNet,I think this might be normal?

hefangnan avatar Oct 15 '23 04:10 hefangnan

It's normal.

Yixuan423 avatar Dec 12 '23 04:12 Yixuan423