UPANets

Reported on 4 benchmarks across 1 task · 1 paper · 2 SOTA

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

Computer Vision4 results

Image ClassificationonTiny-ImageNet
Top 1 Accuracy· 2021-03-15
67.67
SOTA
UPANets: Learning from the Universal Pixel Attention Networks arXiv:2103.08640
Image ClassificationonTiny ImageNet Classification
Validation Acc· 2021-03-15
67.67
best: 92.98 (Astroformer)
SOTA
UPANets: Learning from the Universal Pixel Attention Networks arXiv:2103.08640
Image ClassificationonCIFAR-10
Percentage correct· 2021-03-15
96.47
best: 99.5 (ViT-H/14)
UPANets: Learning from the Universal Pixel Attention Networks arXiv:2103.08640
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2021-03-15
80.29
best: 96.08 (EffNet-L2 (SAM))
UPANets: Learning from the Universal Pixel Attention Networks arXiv:2103.08640