HobbitLong CMC issues - Githubissues

HobbitLong / CMC

[arXiv 2019] "Contrastive Multiview Coding", also contains implementations for MoCo and InstDis

BSD 2-Clause "Simplified" License

1.3k stars 179 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

CMC for small custom dataset

#72 HaoliangZhou opened 2 years ago
0
Pre-trained model for Video task

#71 Shankhanil006 opened 2 years ago
1
How can it be used for more than two views datasets?

#70 A11en0 opened 2 years ago
0
Questions about NCEAverage.py

#69 LourisXu opened 2 years ago
0
question about input image

#68 SkyWalker2099 opened 2 years ago
0
Modified AlexNet architecture for STL-10 dataset

#67 medric49 closed 2 years ago
0
Protential memory leak when training?

#66 Redaimao opened 2 years ago
0
how to use the pretrained MoCo

#65 guomanshan opened 3 years ago
0
Feature Tensor Shape Question

#64 QiyaoWei closed 3 years ago
1
pretrained model of ResNet50v1

#63 lzyhha opened 3 years ago
0
Training Loss Doesn't Improve for CIFAR10

#62 QiyaoWei closed 3 years ago
0
question about shuffle_ids

#61 Nusselder9 opened 3 years ago
0
Create LICENSE

#60 HobbitLong closed 3 years ago
0
About the training epochs for ImageNet-1K

#59 allo-rene opened 3 years ago
0
index drawn by AliasMethod is not on the same gpu as the model

#58 dongyaoli10x closed 4 years ago
1
Format of dataset

#57 zeng-lingyun closed 4 years ago
2
How to prevent positive sample is sampled by AliasMethod?

#56 hschhh666 closed 4 years ago
1
mean std value for LAB/YCbCr normalization

#55 PhilLint opened 4 years ago
1
Gpu memory heavy on gpu0

#54 tankche1 closed 4 years ago
1
Train/Test split when training linear classifier (ImageNet)

#53 jnyjxn opened 4 years ago
3
Weight shape in Linear classifier

#52 bfshi closed 4 years ago
1
Question about data augmentation and memory bank

#51 talshef opened 4 years ago
0
Implementing CMC on CIFAR-10

#50 ShaoTengLiu closed 4 years ago
6
Reproduce MoCov2 on ImageNet 1k

#49 lianqing01 closed 4 years ago
1
Using dot product as a proxy for probability in NCEAverage

#48 vinsis closed 4 years ago
4
Pre-trained weights for linear classifier available?

#47 rgeirhos closed 4 years ago
2
The loss label of of NCESoftmaxLoss in NCECriterion.py?

#46 xuChenSJTU closed 4 years ago
0
A question about the dataset.

#45 xuChenSJTU closed 4 years ago
0
Unable to run pre-trained models

#44 jlindsey15 closed 4 years ago
1
Question about NCECriterion. py

#43 Abyssaledge opened 4 years ago
3
Questions about NCEAverage.py

#42 vinsis closed 4 years ago
5
Reproducing MoCo on ImageNet-1k.

#41 yzou2 closed 4 years ago
5
Question about softmax loss

#40 hyBlue closed 4 years ago
6
shuffle-bn has no effect on single-GPU

#39 sorenrasmussenai opened 4 years ago
1
Expected values of `ins_prob` and `ins_loss` in MoCo when training is working

#38 nlml opened 4 years ago
1
Accuracy on ImageNet using Resnet50v1

#37 semi-supervised-paper opened 4 years ago
0
TenCrop Results

#36 haohang96 closed 4 years ago
1
Unable to reproduce full Imagenet accuracies of pretrained weights for CMC Resnet50v2 and MoCo

#35 ajtejankar closed 4 years ago
5
Augmenting images at the evaluation of downstream classification task

#34 dandelin closed 4 years ago
2
the initial value of memory in Memory Bank of InsDis and CMC

#33 lhcalibur closed 4 years ago
2
Something went wrong when evaluating the results on ImageNet

#32 ChongjianGE opened 4 years ago
1
other available views?

#31 wqtwjt1996 opened 4 years ago
5
Curious about the RandomResizedCrop parameters(minimum crop in your code)

#30 WangFeng18 closed 4 years ago
2
about the imagenet100

#29 kunalmessi10 closed 4 years ago
0
Worse results by reproducing MoCo, InsDIS and CMC on ImageNet100

#28 nkliuyifang closed 4 years ago
5
How to prevent an element in the enqueue come from the same sample as the query?

#27 ZYX-MLer closed 4 years ago
1
During the training, the loss and probs seem bad.

#26 ceezy767 closed 4 years ago
1
ImageNet100

#25 ChongjianGE closed 4 years ago
2
ImageNet100 subset

#24 DJInori closed 4 years ago
2
Estimating normalization factor Z

#23 KaimingHe opened 4 years ago
7