issues
search
HobbitLong
/
SupContrast
PyTorch implementation of "Supervised Contrastive Learning" (and SimCLR incidentally)
BSD 2-Clause "Simplified" License
3.12k
stars
537
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Is there a method that does not require pre training.
#152
Wolongchicken
opened
1 month ago
0
Training using Supcon loss is slower than cross_entropy loss
#151
yinbing668
opened
2 months ago
0
Solving Loss Saturation
#150
avernal1819
opened
2 months ago
0
about loss_in and loss_out in paper
#149
hweejuni
opened
6 months ago
1
How to change backbone
#148
lucasPV
opened
9 months ago
1
solved
#147
limengting-evelyn
closed
10 months ago
0
Code for using the ImageNet pretrained model
#146
LightingMc
opened
10 months ago
4
Moco trick?
#145
LightingMc
opened
10 months ago
0
What mean, std and other datapreprocessing transforms did you use for ImageNet?
#144
LightingMc
opened
10 months ago
1
Pre-trained model
#143
tanushreebanerjee
opened
10 months ago
1
fix: nan loss values
#142
srsawant34
closed
10 months ago
6
NaN loss for no positive pair from SupCon
#141
srsawant34
closed
10 months ago
0
Feature/pre commit
#140
seanswyi
opened
1 year ago
0
Mvp
#139
NinaLong2077
closed
1 year ago
1
.
#138
NinaLong2077
closed
1 year ago
0
Purpose of logits_mask?
#137
adv010
closed
1 year ago
3
CE vs CE + SupCon
#136
qsunyuan
closed
1 year ago
2
loss computation: mean and not sum
#135
CassNot
opened
1 year ago
4
Execution of code for ImageNet code (Moco trick)
#134
Angelina1996
opened
1 year ago
1
file “main_linear.py”， line103， set_model(opt) ，if torch.cuda.is_available() is false， why model will not load_state_dict(state_dict)
#133
wupeng1989
opened
1 year ago
1
Is this using resnet50x4 width?
#132
hdeng26
closed
1 year ago
2
Question about potentially incomplete usage of 'logits_mask'
#131
ruohuali
opened
1 year ago
1
Local pipeline
#130
LukasMahieu
closed
1 year ago
1
File "main_supcon.py", line 270, in main opt = parse_option() File "main_supcon.py", line 94, in parse_option assert opt.degrees is not None AssertionError
#129
noreenanwar
opened
1 year ago
0
the mean and std of cifar dataset
#128
shuaiNJU
opened
1 year ago
0
Loss function is not convergent when batch-sizes smaller？
#127
kyre-99
opened
1 year ago
4
3673 adding image reducer
#126
chrishickey
closed
1 year ago
0
fix comments!
#125
SlowMonk
closed
1 year ago
0
Why use two anchors?
#124
qiaoyu1002
opened
2 years ago
1
The problem about the number of positives in code
#123
shuaiNJU
opened
2 years ago
0
Supervised Contrastive Learning with n_views=1
#122
piconti
opened
2 years ago
1
Using SupContrast loss for non-image application
#121
mossishahi
opened
2 years ago
1
Hyperparameters for ImageNet training?
#120
InugYoon
opened
2 years ago
4
Run custom dataset
#119
geokarant
closed
2 years ago
0
Can anyone reproduce this result or is it just me?
#118
isacyoo
closed
2 years ago
1
MoCo code
#117
carloalbertobarbano
opened
2 years ago
1
About the loss. Sincerely, I would like to ask：
#116
Struggle-Forever
opened
2 years ago
5
Provided pre-trained load issue
#115
Rasoul77
opened
2 years ago
3
compare
#114
alan-minchan-kim
closed
2 years ago
0
About train resent-18 on the dataset CIFAR-100, the accuracy is just 55.
#113
TouchNow
opened
2 years ago
1
f1, f2 = torch.split(features, [bsz, bsz], dim=0)
#112
onlinehuazai
opened
2 years ago
1
fix a bug, which has the little probability of producing nan in the loss
#111
yuanlonghui
opened
2 years ago
2
Question about logits_mask
#110
Alva-2020
opened
2 years ago
5
Another bug?
#109
W2Q3Q1
opened
2 years ago
0
May be a bug?
#108
W2Q3Q1
opened
2 years ago
1
Fixes the usage of SimCLR loss in README.md
#107
vaydemir3
opened
2 years ago
0
SimCLR Loss: False Output (Possibly with Correction)
#106
vaydemir3
opened
2 years ago
2
The implementation using the trick of the momentum encoder
#105
yinguoweiOvO
opened
2 years ago
2
Why the func returns "nan"?
#104
Zhudongsheng75
opened
2 years ago
8
Meaning of joint training
#103
htn274
opened
2 years ago
3
Next