hightower

ok...seems like training from scratch might not be well supported. But i just want to use this repo to verify my network arch, and imagenet pre-trained model is still on...

> Yes, set all parameter to re-trainable seems hard to converge. This year, Mr He did publish a paper named 'Rethinking ImageNet Pre-training' which claimed the pre-train on imagenet is...

> Hi, @1453042287 @cvtower > > I have another issue about the train precision and loss curve. The following is the result from tensorboardX. > >  > > It...

Hello @ruiming46zrm, @twmht The mega test code has to load a trained model and do inference, i mean it can not be tested during training, right? Then i have a...

same error on mac here. however, it works fine in my ubuntu env.

this issue deserves more attention.