rakshith291

Results 2 issues of rakshith291

Interaction between patches through a transpose may have a stronger role to play ?

Hi, I was going through your exp report. You have made a point that since you are able to get a good performance without using attention layer so good performance...

Fails while converting pertained checkpoint from cifar 10

Since in cifar 10, as mentioned in paper we are replacing the filter size and stride in first conv layer of resnet. Probably this is causing the issue , it...