Statements (24)
Predicate | Object |
---|---|
gptkbp:instanceOf |
Vision Transformer variant
|
gptkbp:application |
gptkb:Computer_Vision
Image Classification |
gptkbp:architecture |
gptkb:transformation
|
gptkbp:digestSize |
224x224
|
gptkbp:fullName |
Vision Transformer Tiny
|
gptkbp:hiddenSize |
192
|
https://www.w3.org/2000/01/rdf-schema#label |
ViT-T
|
gptkbp:introduced |
gptkb:An_Image_is_Worth_16x16_Words:_Transformers_for_Image_Recognition_at_Scale
gptkb:Google_Research |
gptkbp:introducedIn |
2020
|
gptkbp:level |
12
|
gptkbp:lowerAward |
gptkb:ViT-B
gptkb:ViT-L gptkb:ViT-S |
gptkbp:MLPSize |
768
|
gptkbp:notableFauna |
3
|
gptkbp:parameter |
5.7 million
|
gptkbp:relatedTo |
gptkb:ViT
|
gptkbp:size |
16x16
|
gptkbp:trainer |
gptkb:ImageNet
|
gptkbp:usedFor |
Efficient vision transformer baselines
|
gptkbp:bfsParent |
gptkb:Vision_Transformer
|
gptkbp:bfsLayer |
6
|