issues
search
yitu-opensource
/
T2T-ViT
ICCV2021, Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
Other
1.14k
stars
177
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Question about the 'unfold operation.
#28
JaminFong
closed
3 years ago
2
Confusion matrix and classify validation images?
#27
zxsitu
closed
3 years ago
1
Nan during training even without '--amp'
#26
Evgeneus
closed
3 years ago
8
LR and weight decay
#25
ZN-Qiao
closed
3 years ago
2
Requirement of ImageNet folder structure and output validation
#24
lcmeng
closed
3 years ago
2
Training Time
#23
blue-blue272
closed
3 years ago
1
Remove classifier head
#22
huixiancheng
closed
3 years ago
2
The performance of only using Performer in the whole network
#21
speedinghzl
closed
3 years ago
2
How to visualize the attention map of t2t-vit?
#20
Salen158
closed
3 years ago
4
NAN Loss for provided model
#19
yix081
closed
3 years ago
8
can it be used for Object Detection?
#18
coderHI328K
closed
3 years ago
1
How to load the provided model.pt.tar to train on my own dataset?
#17
shinianzhihou
closed
3 years ago
1
visualization of resnet50 features
#16
ming1993li
closed
3 years ago
1
Reproduce failure, test accuracy is 0
#15
valuefish
closed
3 years ago
2
The minor difference between Token_transformer class and transformer_block.Block class
#14
miraclebiu
closed
3 years ago
1
the input channels
#13
DDxk369
closed
3 years ago
1
Very low performance within the first 10 epochs
#12
PkuRainBow
closed
3 years ago
3
Token performer not implemented forward method.
#11
imyhxy
closed
3 years ago
1
Cannot get the reported MACs in paper
#10
happywu
closed
3 years ago
5
显存随着训练的进行不断增长
#9
shinianzhihou
closed
3 years ago
5
Only use the first patch after T2T-ViT Backbone
#8
JingXfei
closed
3 years ago
1
I have a problem about input size ?
#7
muzishen
closed
3 years ago
1
Hi, what is the version of torch used?
#6
violet998
closed
3 years ago
0
How to calculate MACs?
#5
pengzhiliang
closed
3 years ago
2
forward
#4
YOU-TO-BE
closed
3 years ago
2
RuntimeError: invalid argument 5: k not in range for dimension at /pytorch/aten/src/THC/generic/THCTensorTopK.cu:26
#3
DDxk369
closed
3 years ago
2
The released models seem broken and cannot be opened?
#2
QiushiYang
closed
3 years ago
3
No forward with Token_performer?
#1
GuideWsp
closed
3 years ago
3
Previous