List your company
ResearchFormerly WizWand

Classification on CheXpert

0.8759AUCDiRAMoCo-v2
0.5627560.6440530.725350.806647Apr 21, 2022Aug 15, 2022Dec 10, 2022Apr 5, 2023Jul 31, 2023Nov 24, 2023Mar 20, 2024
Updated 8mo ago

Evaluation results

MethodLinks
DiRAMoCo-v2Pretraining Dataset=ChestX-ray14, Backbone=ResNet-50, Evaluation Protocol=Fine-tuning2022.040.8759
DiRABarlow TwinsPretraining Dataset=ChestX-ray14, Backbone=ResNet-50, Evaluation Protocol=Fine-tuning2022.040.875
SupervisedPretraining Dataset=ChestX-ray14, Backbone=ResNet-50, Evaluation Protocol=Fine-tuning2022.040.874
SupervisedPretraining Dataset=ImageNet, Backbone=ResNet-50, Evaluation Protocol=Fine-tuning2022.040.8717
RandomPretraining Dataset=None, Backbone=ResNet-50, Evaluation Protocol=Fine-tuning2022.040.8662
KoBo-VitEvaluation Protocol=Few-shot-Frozen, Backbone=ViT, Training data=1%2023.070.866
GoGModel=DeiT-B2024.030.8644
KoBoEvaluation Protocol=Few-shot-Frozen, Backbone=ResNet-50, Training data=1%2023.070.8628
FGG(best)Model=DeiT-B2024.030.8619
MGCAEvaluation Protocol=Few-shot-Frozen, Backbone=ResNet-50, Weights=Reproduced, Training data=1%2023.070.8616
DiRAsimSiamPretraining Dataset=ChestX-ray14, Backbone=ResNet-50, Evaluation Protocol=Fine-tuning2022.040.8604
ConVIRTEvaluation Protocol=Few-shot-Frozen, Backbone=ResNet-50, Weights=Reproduced, Training data=1%2023.070.8548
GloriaEvaluation Protocol=Few-shot-Frozen, Backbone=ResNet-50, Weights=Reproduced, Training data=1%2023.070.8492
GoGModel=ResNet502024.030.8488
GS(best)Model=ResNet502024.030.8444
Greedy SoupModel=ResNet502024.030.8444
FGG(best)Model=ResNet502024.030.8434
GoUModel=ResNet502024.030.8351
MedCLIPEvaluation Protocol=Few-shot-Frozen, Backbone=ResNet-50, Weights=Official, Training data=1%2023.070.8214
GS(best)Model=DeiT-B2024.030.8143
Greedy SoupModel=DeiT-B2024.030.8068
GoUModel=DeiT-B2024.030.7609
Uniform SoupModel=DeiT-B2024.030.7177
Uniform SoupModel=ResNet502024.030.5752
CLIPEvaluation Protocol=Few-shot-Frozen, Backbone=ResNet-50, Weights=Official, Training data=1%2023.070.5748