issues
search
keyu-tian
/
SparK
[ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"
https://arxiv.org/abs/2301.03580
MIT License
1.41k
stars
82
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
downstream_imagenet fine-tuning question
#33
alskdjfasdfsadf
closed
1 year ago
1
What should i do when input shape is a rectangle ?
#32
syjabc
closed
1 year ago
2
how to convert sparse model to dense model?
#31
Mike575
closed
1 year ago
2
Optimized sparse conv from open3d
#30
ifeherva
closed
1 year ago
2
Can you provide more model pth with decoder?
#29
WwwwYz666
closed
1 year ago
2
The finetuning weights on imagenet
#28
YuetongFang
opened
1 year ago
1
The finetuning hyperparameters of resnet50
#27
Vickeyhw
opened
1 year ago
7
cosine learning rate annealing
#26
shuuchen
closed
1 year ago
1
a small model such as Mobilenet v2 for pre-training
#25
mmmz28
opened
1 year ago
3
Will pretrained models converge faster on downstream tasks than non pretrained models?
#24
xylcbd
closed
1 year ago
1
The time of training ConvNext-B
#23
kalelpark
closed
1 year ago
1
Single GPU
#22
hasayake007
closed
1 year ago
1
About Datasets
#20
songbingyue
closed
1 year ago
1
torch.distributed.elastic.multiprocessing.api.SignalException
#19
shuuchen
closed
1 year ago
2
Weights of Resnet with decoder
#18
Vickeyhw
closed
1 year ago
4
How to add support for more custom models in pre-training
#17
zero0kiriyu
closed
1 year ago
3
Is it possible to support the unsquare input
#16
zero0kiriyu
closed
1 year ago
1
Lr layer decay
#15
Vickeyhw
closed
1 year ago
1
vis_reconstruction.py question
#14
buxuewushu1314
closed
1 year ago
6
请问下ppt中的动图是怎么画的
#13
whuhit
closed
1 year ago
2
should sp_conv also multipies 1/ratio like the dropout in training mode?
#12
qianyizhang
closed
1 year ago
1
How to try other mask patch size like 16*16?
#11
rangek
closed
1 year ago
5
Performance gain of sparse convolution compared to zero-outing?
#10
rayleizhu
closed
1 year ago
6
The plan to release the visualization code of image reconstruction and “mask pattern vanishing” issue
#9
songkq
closed
1 year ago
5
The time of training resnet50
#8
ModulatedConvolutionalNetworks
closed
1 year ago
2
q
#7
C929-x
closed
1 year ago
1
Finetuning links broken
#6
R4ZZ3
closed
1 year ago
1
pretrained model on imagenet
#5
AMNAALMGLY
closed
1 year ago
1
When will the full code be released?
#2
zhangzjn
closed
1 year ago
4
Where is the implementation of sparse convolution please?
#1
CoinCheung
closed
1 year ago
9
Previous