List your company
ResearchFormerly WizWand

Fine-grained Image Classification on OxfordPets (test)

4Top-1 ErrorInCA
3.8844.6675.456.233Mar 7, 2023
Updated 8mo ago

Evaluation results

MethodLinks
InCABackbone=ViT-L/162023.034———
Full FTBackbone=ViT-L/162023.034.2———
InCA (last)Backbone=ViT-L/162023.034.2———
LoRABackbone=ViT-L/162023.034.3———
Full FTBackbone=SWIN-L, % Trainable parameters=100%, No backbone backpropagation=false2023.034.6———
InCABackbone=SWIN-L, % Trainable parameters=3.7%, No backbone backpropagation=true2023.034.7———
AdaLNBackbone=SWIN-L, % Trainable parameters=0.1%, No backbone backpropagation=false2023.034.8———
BitFitBackbone=SWIN-L, % Trainable parameters=0.1%, No backbone backpropagation=false2023.034.9———
In. LPBackbone=SWIN-L, % Trainable parameters=0.1%, No backbone backpropagation=true2023.035———
LoRABackbone=SWIN-L, % Trainable parameters=0.8%, No backbone backpropagation=false2023.035.2———
In. MLP-3Backbone=ViT-L/162023.035.3———
BitFitBackbone=ViT-L/162023.035.3———
MLP-3Backbone=ViT-L/162023.035.5———
AdaLNBackbone=ViT-L/162023.035.5———
InCA (last)Backbone=SWIN-L, % Trainable parameters=3.7%, No backbone backpropagation=true2023.035.5———
LPBackbone=SWIN-L, % Trainable parameters=0.1%, No backbone backpropagation=false2023.035.5———
In. MLP-3Backbone=SWIN-L, % Trainable parameters=2.8%, No backbone backpropagation=true2023.035.5———
MLP-3Backbone=SWIN-L, % Trainable parameters=2.8%, No backbone backpropagation=false2023.035.5———
In. LPBackbone=ViT-L/162023.036.1———
LPBackbone=ViT-L/162023.036.4———
VPTBackbone=ViT-L/162023.036.9———
CLIPBackbone=ViT-B/16, Shots=0, Context Length=manual prompt2022.03—91.1797.2694.12
CoCoOpBackbone=ViT-B/16, Shots=16, Context Length=42022.03—95.297.6996.43
CoOpBackbone=ViT-B/16, Shots=16, Context Length=42022.03—93.6795.2994.47