Collaboration of Experts: Achieving 80% Top-1 Accuracy on ImageNet with 100M FLOPs

Open paper
Benchmark Model Rank Results
image-classification-on-imagenetCoE-Large–Top 1 Accuracy: 80.7%Number of params: 95.3MGFLOPs: 0.194
image-classification-on-imagenetCoE-Large + CondConv–Top 1 Accuracy: 81.5%GFLOPs: 0.214
image-classification-on-imagenetCoE-Small + CondConv + PWLU–Top 1 Accuracy: 80%GFLOPs: 0.100

Rank counts only results with a code link.