kekmodel FixMatch-pytorch issues

kekmodel / FixMatch-pytorch

Unofficial PyTorch implementation of "FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence"

MIT License

758 stars 170 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

训练的准确率一直是10%，显然是没有作用的因为一共就只有10个类别

#54 hanyiwei799 opened 2 months ago
0
每次加载训练好的模型，结果都不一样

#53 onlinehuazai opened 8 months ago
0
RuntimeError: "nll_loss_forward_reduce_cuda_kernel_2d_index" not implemented for 'Int'

#52 wxqyyds opened 8 months ago
0
Poor Results on CIFAR-100 with Smaller Models (e.g., ResNet34, ResNet18)

#51 tsunghan-wu opened 1 year ago
0
why add a interleave before the input?

#50 williamium3000 closed 10 months ago
3
Can't reproduce the result with single GPU

#49 moucheng2017 opened 2 years ago
0
All labeled examples are as part of unlabeled data without using their labels

#48 chengerr closed 2 years ago
1
FixMatch for Multi-class Classification

#47 GuoJingtao-1997 closed 2 years ago
0
Why the batch_size of CIFAR-100 set to 64 (total number over 4 GPU) ?

#46 kevinbro96 closed 2 years ago
0
Training time without distributed training?

#45 07Agarg closed 2 years ago
2
v divided by 2 problem in Cutout

#44 Fordacre closed 2 years ago
1
No Validation Set

#43 07Agarg closed 2 years ago
1
train failed when EMA mode is off

#42 frozenzo opened 3 years ago
5
Lack of seed during random labeled image selection may be leading to better performance on training resumption

#41 dsouzinator closed 3 years ago
5
Wrong use of AverageMeter

#40 huberl closed 3 years ago
1
The RandAugment Implementation Is Wrong

#39 psellcam closed 2 years ago
3
Performance not reproduced

#38 Hyun1A closed 2 years ago
0
Invalid syntax in train.py

#37 mdziezyc closed 2 years ago
1
Purpose of Interleave

#36 feras-oughali closed 3 years ago
3
total numer of iterations

#35 holzbock opened 3 years ago
2
set_epoch with DistributedSampler

#34 TonyLianLong closed 3 years ago
1
RandAugment Implementation

#33 psellcam closed 2 years ago
3
What is the purpose of the parameter 'activate_before_residual' in WRN?

#32 KittyStore closed 2 years ago
0
How can I train model using multi GPUS in a machine?

#31 Kyfafyd closed 3 years ago
7
Have you tried SVHN dataset?

#30 fiona-lxd closed 2 years ago
3
Can not get repo reported accuracy for cifar10@40.5

#29 Kyfafyd closed 3 years ago
5
Have you fetch labelled samples for each class in balance?

#28 Kyfafyd closed 3 years ago
2
Accuracy lower than reported

#27 fil-mp closed 3 years ago
2
RuntimeError after 1 epoch

#26 valeriydutt closed 3 years ago
2
Why the result with this implementation is better than reported one in the original paper?

#25 sooonwoo opened 3 years ago
2
When fix patch is applied to my dataset, the dataset is unbalanced. Why does the training set perform well, but the loss of verification set first decreases and then keeps rising?

#24 a571-op opened 3 years ago
3
Using fixmatch to train on my dataset, LR = 0.03, batchsize = 64. When the unlabeled loss increases gradually, the model gets worse and worse, and the loss of training set will decrease, but the loss of verification set will increase, and the accuracy will decrease?

#23 a571-op closed 2 years ago
1
How to delete this question, I've solved it

#22 ghost closed 3 years ago
0
fail to reproduce the results of your experiment

#21 Jiawen-huang closed 3 years ago
1
Datasets

#20 fil-mp closed 3 years ago
2
What does the function `interleave` do?

#19 charlesCXK closed 4 years ago
4
unlabeled data contain the labeled data?

#18 yaojunr closed 4 years ago
1
custom dataset

#17 KirillHiddleston closed 4 years ago
1
PyTorch Distributed Training

#16 ZhuoranYu closed 4 years ago
1
Weight decay in EMA?

#15 bkj closed 4 years ago
2
Accuracy curves

#14 bkj closed 4 years ago
1
CIFAR100 has different architecture from official implementation

#13 zijian-hu closed 4 years ago
1
Any plans to add CTAugment?

#12 monkeysforever closed 3 years ago
2
Unsupervised loss part for one single class

#11 dendrobiumz closed 4 years ago
1
All labeled examples are as part of unlabeled data without using their labels?

#10 Chen-Song closed 4 years ago
6
1

#9 Chen-Song closed 4 years ago
0
Cannot reproduce 94.72% Top-1 accuracy on CIFAR10 (4000 labeled data)

#8 tqtrunghnvn closed 4 years ago
15
BatchNorm has a biased estimation

#7 YUE-FAN closed 3 years ago
4
cos learing rate

#6 YUE-FAN closed 4 years ago
2
num_labeled in DistributedDataParallel

#5 LiheYoung opened 4 years ago
9