e_yi

typo

1

https://github.com/vandit15/Class-balanced-loss-pytorch/blob/921ccb8725b1eb0903b2c22a1a752a594fcae138/class_balanced_loss.py#L28 should be `alpha: A float tensor of size [num_classes]`

shareWrr.mat seems to be wrong

1

`newWrr` stored in shareWrr.mat does not equal to `Wrd*Wdr`.

## 🚀 Feature support saving/loading batched graphs (by preserving the information for unbatching during saving). ## Motivation It would greatly imporve the time efficiency (and maybe space too?) comparing with...

ERR_EMPTY_RESPONSE 日志显示System.NullReferenceException

3

版本v5.1.5 windows10 预览版 版本号1903 版本18950.1000 log如下 >[2019-08-05 23:50:03] [Info] Connect lh3.googleusercontent.com:443 via hgc.wmsxwd.xyz:10091 [2019-08-05 23:50:03] [Info] DNS nolock query lh3.googleusercontent.com answer 172.217.24.1 [2019-08-05 23:50:03] [Error] System.NullReferenceException: 未将对象引用设置到对象的实例。 在 Shadowsocks.Util.Utils.QueryDns(String host,...

The units in the performance table are not consistent.

1

The units in the performance table are not consistent. For example, `0.0246` should be `24.6` ?

Support A Variable Number of Batches

1

### Description & Motivation I have a customized batch sampler that has an undetermined number of batches (to maximize the use of GPU memory with variable-sized samples). I believe this...