hexin6688
Results
4
comments of
hexin6688
when trainning on the first stage,the val loss=nan,but when training on the second stage(unfreezing) the val loss starts to descending.
> > > 解决办法就是加载越训练模型 哈哈 我怎么不能加载预训练模型呢？checkpoint里面有需要的demo.ckpt
anybody soveled this issue?
hi！if I want to use the 'Cooratt' in resnet50, should I replace the 'RELU6' to 'relu'?