syoya

Thanks for sharing such a good framework. I'm curious that when I'm using WideResNet with depth-k (26-10), it actually has less GPU usage than DenseNet-BC(k=40) even though WideResNet has more...

Nan loss for quantization

5

I'm running the code for 8-bit quantization but found that the training loss always gets NAN while I didn't make a slight modification to the original code. Wondering why this...

Config file of CIFAR10 experiments

9

Thanks for your work. I've tried to modify the current config files of fixmatch-ccssl on CIFAR100 and failed to reproduce the same reuslts on CIFAR10 as your paper shows. Would...

Did you use warmup technique?

1

Thanks for your sharing. I've noticed that the generalization gap between small batch size and large batch size is not the same as described as in the original paper. I'm...

Thanks for your great work and sharing. I've noticed that the training code and dataset would be published after Apr 20 listed in your roadmap. Is there any problem causing...

full cpu usage

4

When I add `nn.DataParallel` into the code so that I can use multiple gpus with default worker num, I find that all my cpus are fully loaded. I haven't check...

Thanks for your effort and great work. I finished stage1 and stage2 training using original animate anyone architecture but found out that the inference code is mainly for the hacked...