ConvMLP: Hierarchical Convolutional MLPs for Vision

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-10ConvMLP-M#36Percentage correct: 98.6
image-classification-on-cifar-10ConvMLP-L#37Percentage correct: 98.6
image-classification-on-cifar-10ConvMLP-S#61Percentage correct: 98
image-classification-on-cifar-100ConvMLP-M#34Percentage correct: 89.1
image-classification-on-cifar-100ConvMLP-L#35Percentage correct: 88.6
image-classification-on-cifar-100ConvMLP-S#46Percentage correct: 87.4
image-classification-on-flowers-102ConvMLP-S#9Accuracy: 99.5
image-classification-on-flowers-102ConvMLP-L#10Accuracy: 99.5
image-classification-on-imagenetConvMLP-L#679Top 1 Accuracy: 80.2%Number of params: 42.7M
image-classification-on-imagenetConvMLP-M#751Top 1 Accuracy: 79%Number of params: 17.4M
image-classification-on-imagenetConvMLP-S#843Top 1 Accuracy: 76.8Number of params: 9M
semantic-segmentation-on-ade20kConvMLP-L#210Validation mIoU: 40
semantic-segmentation-on-ade20kConvMLP-M#211Validation mIoU: 38.6
semantic-segmentation-on-ade20kConvMLP-S#217Validation mIoU: 35.8

Rank counts only results with a code link.