GFNet-H-B

Reported on 5 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision5 results

Image ClassificationonStanford Cars
Accuracy· uses extra data· 2021-07-01
93.2
best: 96.868 (efficient adaptive ensembling)
Global Filter Networks for Image Classification arXiv:2107.00645
Image ClassificationonCIFAR-10
Percentage correct· uses extra data· 2021-07-01
99
best: 99.5 (ViT-H/14)
Global Filter Networks for Image Classification arXiv:2107.00645
Image ClassificationonFlowers-102
Accuracy· uses extra data· 2021-07-01
98.8
best: 99.76 (CCT-14/7x2)
Global Filter Networks for Image Classification arXiv:2107.00645
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2021-07-01
90.3
best: 96.08 (EffNet-L2 (SAM))
Global Filter Networks for Image Classification arXiv:2107.00645
Image ClassificationonImageNet
GFLOPs· 2021-07-01
8.6
best: 1478 (InternImage-H)
Global Filter Networks for Image Classification arXiv:2107.00645