CoinCheung/pytorch-loss issues and pull requests

#41 - should use log_probs in affinity loss

Pull Request - State: open - Opened by eliqbiq 25 days ago - 1 comment

#40 - hard triplet conergence

Issue - State: open - Opened by ffredd about 2 years ago

#39 - installation failed

Issue - State: open - Opened by mosheliv about 2 years ago - 3 comments

#38 - “THC/THC.h”: No such file or directory

Issue - State: open - Opened by cop123123 over 2 years ago - 5 comments

#37 - remove THC.h

Pull Request - State: closed - Opened by zyddnys over 2 years ago - 1 comment

#36 - fatal error: THC/THC.h: No such file or directory

Issue - State: closed - Opened by zyddnys over 2 years ago - 2 comments

#35 - some question in rankloss

Issue - State: closed - Opened by nijiazhi over 2 years ago - 5 comments

#34 - class balanced loss?

Issue - State: open - Opened by 2533245542 over 2 years ago - 1 comment

#33 - partial_fc_amsoftmax memory leak

Issue - State: closed - Opened by tianxingyzxq almost 3 years ago - 5 comments

#32 - Regularizing Neural Networks by Penalizing Confident Output Distribution

Issue - State: closed - Opened by eakirtas almost 3 years ago - 4 comments

#31 - Amazing work！

Issue - State: closed - Opened by anxiangsir almost 3 years ago - 1 comment

#30 - LabelSmoothSoftmaxCEV1, LabelSmoothSoftmaxCEV2, LabelSmoothSoftmaxCEV3

Issue - State: closed - Opened by quanweiliu almost 3 years ago - 3 comments

#29 - Question about Multi-class focal loss

Issue - State: closed - Opened by DingYuan0118 almost 3 years ago - 4 comments

#28 - fatal error: math.h: No such file or directory

Issue - State: open - Opened by snakers4 about 3 years ago - 2 comments

#27 - your focal loss is wrong? it seems little different with others, can you explain your code?

Issue - State: closed - Opened by Jack-zz-ze about 3 years ago - 8 comments

#26 - Calculation about focal loss

Issue - State: closed - Opened by BinsonW about 3 years ago - 1 comment

#25 - 哪一个损失函数最好用，损失函数类中里面又不带参数？

Issue - State: closed - Opened by jingzhengli about 3 years ago - 1 comment

#24 - TPU support

Issue - State: closed - Opened by talhaanwarch about 3 years ago - 1 comment

#23 - use of TO_REMOVE in GIOU loss

Issue - State: closed - Opened by souvik3333 over 3 years ago - 1 comment

#22 - there is some problems in the file of focal_loss.py

Issue - State: closed - Opened by OliviaDdy over 3 years ago - 4 comments

#21 - LovaszSoftmax Loss test case not working

Issue - State: closed - Opened by omerferhatt over 3 years ago - 4 comments

#20 - 如何调用tripletloss？

Issue - State: closed - Opened by 392398741nm over 3 years ago - 1 comment

#19 - The probability caculation of negative class in 'Label smooth.py'

Pull Request - State: open - Opened by kaijieshi7 over 3 years ago

#18 - feature request: dice loss support multi-class classification ?

Issue - State: closed - Opened by feng-1985 over 3 years ago - 1 comment

#17 - new feature request: dice loss

Issue - State: closed - Opened by feng-1985 over 3 years ago

#16 - 请问am-softmax可以用在多分类任务上么，比如网络结构最后一层输出分类是7，那amsoftmax(）中的参数填啥，第一个in_feats应该填7，第二个num_class也填7,这应该不对，不知道怎么使用，请教了

Issue - State: closed - Opened by wbb95 over 3 years ago - 2 comments

#15 - Any plan to give a tutorial?

Issue - State: closed - Opened by seanzhuh over 3 years ago - 1 comment

#14 - need cuda for installing on MacOs

Issue - State: closed - Opened by reymondzzzz over 3 years ago - 1 comment

#13 - where is the mish_cpp?

Issue - State: closed - Opened by hhl0202 almost 4 years ago - 3 comments

#12 - I want to use FocalLossV1,but has the error that your loss has no attribute to backward().

Issue - State: closed - Opened by hanna6666 almost 4 years ago - 5 comments

#11 - use on multi-label 3D medical image segmentation task?

Issue - State: closed - Opened by xwjBupt almost 4 years ago - 2 comments

#10 - BCEWithLogitsLoss has combined a Sigmoid layer and the BCELoss in one single class, But why to use torch.sigmoid again

Issue - State: closed - Opened by sunpeng981712364 about 4 years ago - 1 comment

#9 - 关于triplet loss

Issue - State: closed - Opened by AnnaXiong about 4 years ago - 3 comments

#8 - do buffered params in EMA need to be updated?

Issue - State: closed - Opened by DietDietDiet about 4 years ago - 1 comment

#7 - Regarding Generalized IOU Loss

Issue - State: closed - Opened by samyak0210 about 4 years ago - 2 comments

#6 - dual focal loss

Issue - State: closed - Opened by damhurmuller over 4 years ago - 7 comments

#5 - AM-softmax implement details

Issue - State: closed - Opened by sweetTT over 4 years ago - 2 comments

#4 - about smooth label

Issue - State: closed - Opened by mk123qwe over 4 years ago - 1 comment

#3 - How do I design my loss fuction like smoothL1？

Issue - State: closed - Opened by kelisiya over 4 years ago - 8 comments

#2 - What does ignore_lb = 255 do?

Issue - State: closed - Opened by wusaifei about 5 years ago - 1 comment

#1 - The purpose of lb_ignore in LabelSmoothSoftmaxCE?

Issue - State: closed - Opened by LightToYang about 5 years ago - 2 comments

GitHub / CoinCheung/pytorch-loss issues and pull requests