Open ildoonet opened 4 years ago
Dark Blue is for B0 and Sky Blue is for B0 with CondConv. I ran this with the above command.
Sorry, Dark blue is for b1. But, still, B0 with CondConv can not achieve the performance as the original paper claims.
Sorry, Dark blue is for b1. But, still, B0 with CondConv can not achieve the performance as the original paper claims.
Hi, I have met the same problem in CIFAR10. CondConv seems not to work. Do you solve this problem now, thanks!
I use below arguments to train condconv-b0 model, which performs 21.7 top1 error reported in the original paper. But I got poor performance such as 22.67 top 1 error.
Can you provide the specific arguments to reproduce?
Can you provide your training logs, such as tensorboard 'top_1_accuracy' graph?
Thanks.