Spijkervet/SimCLR issues and pull requests

#50 - Add support for MPS backend (Apple Metal)

Pull Request - State: open - Opened by hsadok 4 months ago

#49 - Fix torchvision pretrained deprecation warning

Pull Request - State: open - Opened by jojje 5 months ago

#48 - Deprecation warning noise on each run (torchvision from pretrained)

Issue - State: open - Opened by jojje 5 months ago

#47 - Utility to plot confusion matrix

Pull Request - State: closed - Opened by Madness4000 10 months ago

#46 - Fix mask issue in distributed nt_xent loss

Pull Request - State: closed - Opened by lxysl about 1 year ago

#45 - Checkpoint at epoch 0 will be covered in the end when using DDP

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

#44 - No upscale in image augmentation?

Issue - State: closed - Opened by vedantroy about 2 years ago

#43 - Issue about the NT_Xent loss

Issue - State: closed - Opened by 980202006 about 2 years ago - 1 comment

#42 - Added support for TFIW dataset

Pull Request - State: open - Opened by mishragauravgm about 2 years ago - 1 comment

#41 - Unify models

Pull Request - State: closed - Opened by JSchweisthal over 2 years ago

#40 - Some questions about the projection head.

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

#39 - Loss decerases slowly in LFW dataset, Is it normal?

Issue - State: open - Opened by muyuuuu almost 3 years ago

#38 - Question about multi-gpu training

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

#37 - Multi GPU support is ... ornamental?

Issue - State: closed - Opened by mbabadi about 3 years ago - 7 comments

#36 - Features computed in def inference(loader, simclr_model, device) in linear_evaluation.py are changing everytime if we run the script again

Issue - State: closed - Opened by yang-ruixin about 3 years ago

#35 - Mistakes in linear_evaluation.py

Issue - State: closed - Opened by q642936448 about 3 years ago

#34 - exclude_from_weight_decay added

Pull Request - State: open - Opened by sadimanna about 3 years ago - 6 comments

#33 - Code explanation in gather.py

Issue - State: open - Opened by littleredxh over 3 years ago - 1 comment

#32 - Is simclr's package model trained on cifar-10?

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

#31 - Hello, If I want to use vgg16, how can I modify it?

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

#30 - The loss doesn't decrease when using multi nodes.

Issue - State: open - Opened by lailvlong over 3 years ago - 10 comments

#29 - Fix TypeError in linear_evaluation.py

Pull Request - State: closed - Opened by robbiejones96 over 3 years ago

#28 - I got some problems when I was testing

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

#27 - Maybe a small bug in LARS implementation

Issue - State: open - Opened by Big-Brother-Pikachu over 3 years ago

#26 - update distributed sampler

Pull Request - State: closed - Opened by steermomo over 3 years ago

#25 - normalize is not used in data augmentation

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

#24 - Hi spijkervet, when youreproduce the experiment on CIFAR, what is the image size and batch size you set?

Issue - State: closed - Opened by LanXiaoPang613 almost 4 years ago - 1 comment

#23 - How to train a model in the same terminal when I used DDP training？

Issue - State: closed - Opened by szhang963 almost 4 years ago - 3 comments

#22 - What is world_size in this case? NT_Xent seems to ask this as a default argument but it hasn't been provided in the code?

Issue - State: closed - Opened by ajaykrishnan23 about 4 years ago - 2 comments

#21 - fixed modules references

Pull Request - State: closed - Opened by Serhiy-Shekhovtsov about 4 years ago - 1 comment

#20 - Fixed no encoding specified

Pull Request - State: open - Opened by ghost about 4 years ago

#19 - Logistic Regression Issue without "eval()"

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

#18 - Loss function is giving different results from tensorflow implementation

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

#17 - Update simclr.py

Pull Request - State: closed - Opened by gauenk about 4 years ago

#16 - Slightly different strength parameter in CIFAR-10 setting

Issue - State: closed - Opened by nzw0301 about 4 years ago - 1 comment

#15 - Why L2 norm then CosSim?

Issue - State: closed - Opened by BoPang1996 about 4 years ago - 5 comments

#14 - train on Imagenet

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

#13 - inter-gpu communication

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

#12 - CIFAR10 accuracy

Issue - State: open - Opened by MrChenFeng about 4 years ago - 3 comments

#11 - query regarding usage of TPU

Issue - State: closed - Opened by ambekarsameer96 about 4 years ago - 2 comments

#10 - torch version

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

#9 - Use resnet v1 and CIFAR stem

Pull Request - State: closed - Opened by alegonz over 4 years ago

#8 - Mismatching resnet50 model for CIFAR-10 experiments

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

#7 - Issue loading mask_correlated_samples

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

#6 - Error when loading the datasets

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

#5 - Small refactoring of your loss

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

#4 - About the Cifar-10 accuracy

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

#3 - Shouldn't this line be `args.epochs` instead of `len(loader)`?

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

#2 - Create LICENSE

Pull Request - State: closed - Opened by Spijkervet over 4 years ago

#1 - Fixed mask_correlated_samples to return masks

Pull Request - State: closed - Opened by isaaccorley over 4 years ago - 1 comment

GitHub / Spijkervet/SimCLR issues and pull requests