issues
search
jeonsworld
/
ViT-pytorch
Pytorch reimplementation of the Vision Transformer (An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale)
MIT License
1.95k
stars
374
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
About imagenet-21k
#10
zhangzjn
closed
3 years ago
2
How can the checkpoint be continued?
#9
gwang-kim
closed
3 years ago
0
Multi-GPU
#8
kamalkraj
closed
3 years ago
6
Train from scratch
#7
stomachacheGE
closed
3 years ago
1
Why the position_embeddings are zeros?
#6
Erichen911
closed
4 years ago
1
The Encoder implementation is different from the original "Attention is all need" paper?
#5
chaoyanghe
closed
4 years ago
4
About the args.image_size
#4
chaoyanghe
closed
4 years ago
1
Tensors do not match?
#3
chaoyanghe
closed
4 years ago
1
Model Architecture For Fine-tuning
#2
chaoyanghe
closed
4 years ago
11
could you help to provide a weight converter which can convert checking points from official TF-based ViT to PyTorch?
#1
chaoyanghe
closed
4 years ago
5
Previous