facebookresearch/barlowtwins issues and pull requests

#56 - Why c.div_(self.args.batch_size) is needed?

Issue - State: closed - Opened by mongeoroo 10 months ago - 1 comment

#55 - Pre-training model for CIFAR

Issue - State: open - Opened by LiYunJamesPhD over 1 year ago

#55 - Pre-training model for CIFAR

Issue - State: open - Opened by LiYunJamesPhD over 1 year ago

#54 - providing the linear ImageNet classifier weights

Issue - State: open - Opened by nikparth over 1 year ago

#54 - providing the linear ImageNet classifier weights

Issue - State: open - Opened by nikparth over 1 year ago

#53 - NaN's introduced during training.

Issue - State: open - Opened by charliebudd over 1 year ago

#53 - NaN's introduced during training.

Issue - State: open - Opened by charliebudd over 1 year ago

#52 - efficiency proposal

Issue - State: closed - Opened by joihn over 1 year ago - 1 comment

#52 - efficiency proposal

Issue - State: closed - Opened by joihn over 1 year ago - 1 comment

#51 - Error in saving resnet50.pth

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

#50 - Possible bug on the loss computation

Issue - State: closed - Opened by azhenone almost 2 years ago - 1 comment

#49 - Quality of Embeddings

Issue - State: open - Opened by ThomasNorr almost 2 years ago

#49 - Quality of Embeddings

Issue - State: open - Opened by ThomasNorr almost 2 years ago

#48 - Question about Fig. 4 in the paper

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

#47 - Where could we find the "reproduced version" of the other SSL methods

Issue - State: closed - Opened by HuangChiEn almost 2 years ago - 1 comment

#47 - Where could we find the "reproduced version" of the other SSL methods

Issue - State: closed - Opened by HuangChiEn almost 2 years ago - 1 comment

#46 - Applications on one-dimensional signal datasets

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

#46 - Applications on one-dimensional signal datasets

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

#45 - weight update for step=0,epoch=0 get's missed.

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

#44 - Start index for each epoch

Issue - State: closed - Opened by toannguyen1904 almost 2 years ago - 3 comments

#44 - Start index for each epoch

Issue - State: closed - Opened by toannguyen1904 almost 2 years ago - 3 comments

#43 - A question on the BT loss with Batch Norm layers

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

#43 - A question on the BT loss with Batch Norm layers

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

#42 - Add dlib implementation to the community updates

Pull Request - State: open - Opened by arrufat over 2 years ago - 2 comments
Labels: CLA Signed

#42 - Add dlib implementation to the community updates

Pull Request - State: open - Opened by arrufat over 2 years ago - 2 comments
Labels: CLA Signed

#41 - Dose we can add distorted image, like Y_a, Y_b, Y_c?

Issue - State: open - Opened by dadadahei over 2 years ago

#40 - About the last normalization layer

Issue - State: closed - Opened by bluenights over 2 years ago - 1 comment

#40 - About the last normalization layer

Issue - State: closed - Opened by bluenights over 2 years ago - 1 comment

#39 - When using the own dataset, loss is around 10,000

Issue - State: closed - Opened by yzakzero over 2 years ago - 8 comments

#39 - When using the own dataset, loss is around 10,000

Issue - State: closed - Opened by yzakzero over 2 years ago - 8 comments

#38 - Will BarlowTwins overfit on the training data?

Issue - State: open - Opened by cucutone over 2 years ago

#38 - Will BarlowTwins overfit on the training data?

Issue - State: open - Opened by cucutone over 2 years ago

#37 - Augmentation Distribution

Issue - State: closed - Opened by gabrieldernbach over 2 years ago - 1 comment

#37 - Augmentation Distribution

Issue - State: closed - Opened by gabrieldernbach over 2 years ago - 1 comment

#36 - Issue loading checkpoint.pth file

Issue - State: closed - Opened by davidrzs over 2 years ago - 1 comment

#36 - Issue loading checkpoint.pth file

Issue - State: closed - Opened by davidrzs over 2 years ago - 1 comment

#35 - transforms.Solarize probability is zero for transforms but not transforms_prime

Issue - State: closed - Opened by dewball345 over 2 years ago - 3 comments

#34 - Master h clip long files

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

#33 - Why do we average out correlation matrices from different GPUs? Is this mathematically valid?

Issue - State: closed - Opened by radekd91 over 2 years ago - 4 comments

#32 - How to reproduce VOC07 linear classification result?

Issue - State: open - Opened by yxchng over 2 years ago

#32 - How to reproduce VOC07 linear classification result?

Issue - State: open - Opened by yxchng over 2 years ago

#31 - Implementation question

Issue - State: closed - Opened by zhangdan8962 almost 3 years ago - 1 comment

#30 - Batchnorm1d

Issue - State: closed - Opened by WesleyHsieh0806 almost 3 years ago

#30 - Batchnorm1d

Issue - State: closed - Opened by WesleyHsieh0806 almost 3 years ago

#29 - Pretrained weight for predictor

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

#28 - how to perform correlation for 4d tensor.

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

#28 - how to perform correlation for 4d tensor.

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

#27 - May I know how to implement BarlowTwins in Segmentation task? Thanks a lot.

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

#27 - May I know how to implement BarlowTwins in Segmentation task? Thanks a lot.

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

#26 - BT loss value on val set every N training epochs without classifier

Issue - State: closed - Opened by neerajwagh almost 3 years ago - 6 comments

#26 - BT loss value on val set every N training epochs without classifier

Issue - State: closed - Opened by neerajwagh almost 3 years ago - 6 comments

#25 - use barlowtwins predict but blocked and can not be killed

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

#25 - use barlowtwins predict but blocked and can not be killed

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

#24 - Improvement over filtering bias and bn

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

#24 - Improvement over filtering bias and bn

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

#23 - Can we ignore the additional parameter --scale-loss?

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

#23 - Can we ignore the additional parameter --scale-loss?

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

#22 - DDP Wrapper in evaluate.py

Issue - State: closed - Opened by zhangdan8962 about 3 years ago - 1 comment

#21 - loss is NaN for some batches/steps

Issue - State: closed - Opened by neerajwagh about 3 years ago

#21 - loss is NaN for some batches/steps

Issue - State: closed - Opened by neerajwagh about 3 years ago

#20 - Evaluating a checkpoint

Issue - State: closed - Opened by kartikgupta-at-anu about 3 years ago - 1 comment

#20 - Evaluating a checkpoint

Issue - State: closed - Opened by kartikgupta-at-anu about 3 years ago - 1 comment

#19 - Typo in the architecture image

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

#19 - Typo in the architecture image

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

#18 - Training loss abruptly changes

Issue - State: closed - Opened by Mushtaqml about 3 years ago - 2 comments

#18 - Training loss abruptly changes

Issue - State: closed - Opened by Mushtaqml about 3 years ago - 2 comments

#17 - Projector Network

Issue - State: closed - Opened by Donny3000 about 3 years ago - 2 comments

#16 - Barlow Twins loss on identical vector

Issue - State: closed - Opened by ChanLIM about 3 years ago - 6 comments

#16 - Barlow Twins loss on identical vector

Issue - State: closed - Opened by ChanLIM about 3 years ago - 6 comments

#15 - Possible to release projector weights too?

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

#15 - Possible to release projector weights too?

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

#14 - Discrepancy in the semi-supervised results for 10% data

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

#14 - Discrepancy in the semi-supervised results for 10% data

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

#13 - Training on mutiple nodes

Issue - State: closed - Opened by d-li14 about 3 years ago - 6 comments

#13 - Training on mutiple nodes

Issue - State: closed - Opened by d-li14 about 3 years ago - 6 comments

#12 - Does this technique work on medical images ?

Issue - State: closed - Opened by ajoseph12 over 3 years ago - 8 comments

#12 - Does this technique work on medical images ?

Issue - State: closed - Opened by ajoseph12 over 3 years ago - 8 comments

#11 - Bug Issues: loaded state dict has a different number of parameter groups

Issue - State: closed - Opened by xinyu1205 over 3 years ago - 3 comments

#10 - Are negative samples required?

Issue - State: closed - Opened by zishanahmed-tomtom over 3 years ago - 3 comments

#10 - Are negative samples required?

Issue - State: closed - Opened by zishanahmed-tomtom over 3 years ago - 3 comments

#9 - are the parameters for 300 epoch with accuracy 71.4% same to the gitgub 1000 epoch?

Issue - State: closed - Opened by Dongshengjiang over 3 years ago - 7 comments

#9 - are the parameters for 300 epoch with accuracy 71.4% same to the gitgub 1000 epoch?

Issue - State: closed - Opened by Dongshengjiang over 3 years ago - 7 comments

GitHub / facebookresearch/barlowtwins issues and pull requests