facebookresearch/simsiam
PyTorch implementation of SimSiam https//arxiv.org/abs/2011.10566
PythonNOASSERTION
Issues
- 0
Loss increasing
#50 opened by karthik1145 - 0
Could not find any class folder
#49 opened by chandagrover - 2
Why are training and testing so slow
#48 opened by InstantWindy - 1
Slow convergence with SGD linear evaluation
#37 opened by gergopool - 0
which vector is used in experiment (the vector atfer predictor or after projection)?
#47 opened by aruba01 - 0
self.encoder.fc in builder.py
#46 opened by MSutharsan - 4
- 2
l2 normalize
#17 opened by UQ-DongYuan - 1
- 1
- 1
Pre-trained weights for SimCLR, MoCov2, BYOL, SwaV
#43 opened by ninatu - 14
Reproducing CIFAR-10 Results
#9 opened by ardywibowo - 1
Loss collapse during training
#40 opened by aisagarw - 3
- 0
- 4
loss question
#29 opened by daeing - 2
SyncBatchnorm usage in main_lincls.py
#38 opened by myoh97 - 0
checkpoint for epoch 800
#36 opened by elias-ramzi - 0
- 3
- 2
Stuck in the DDP initiliazation step
#16 opened by kxgong - 1
difference between byol in table19 and simsiam?
#34 opened by LMerCy - 3
Single GPU Training
#30 opened by nixczhou - 1
Loss descent slowly
#31 opened by vincezengqiang - 1
Checkpoints with 50 epochs
#32 opened by kevinbro96 - 7
Checkpoint with 200 epochs?
#18 opened by lzyhha - 1
zero_init_residual
#28 opened by ifountalis - 1
- 2
- 2
- 2
- 6
- 1
Checkpoints for COCO and VOC
#19 opened by alexlioralexli - 1
- 1
Negative Loss
#21 opened by say2sarwar - 3
What's the point if we do not gather all outputs in different GPUs to compute contrastive loss
#20 opened by BaohaoLiao - 3
- 5
SyncBatchNorm
#12 opened by dengandong - 1
issue about the cosine similarity loss function
#14 opened by sonnefred - 0
Question about batch size
#15 opened by CoinCheung - 1
- 1
Fixed LR on predictor with LR warmup
#11 opened by patrickpoirson - 2
kNN evaluation during training
#10 opened by elias-ramzi - 1
- 0
teaser
#2 opened by endernewton - 0
Teaser
#1 opened by endernewton