Mixer-B/16- SAM

Reported on 5 benchmarks across 2 tasks · 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 ClassificationonCIFAR-10
Percentage correct· 2021-06-03
97.8
best: 99.5 (ViT-H/14)
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations arXiv:2106.01548
Image ClassificationonFlowers-102
Accuracy· uses extra data· 2021-06-03
90
best: 99.76 (CCT-14/7x2)
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations arXiv:2106.01548
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2021-06-03
86.4
best: 96.08 (EffNet-L2 (SAM))
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations arXiv:2106.01548
Image ClassificationonOxford-IIIT Pets
Accuracy· 2021-06-03
92.5
best: 97.1 (EffNet-L2 (SAM))
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations arXiv:2106.01548
Fine-Grained Image ClassificationonOxford-IIIT Pets
Accuracy· 2021-06-03
92.5
best: 97.1 (EffNet-L2 (SAM))
When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentations arXiv:2106.01548