RepAn: Enhanced Annealing through Re-parameterization

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-100RepAn (ResNet-18, ACNet)#170Percentage correct: 78.37
image-classification-on-cifar-10RepAn (ResNet-18, DBB)#177Percentage correct: 94.68
image-classification-on-imagenetRepAn (ResNet-50)#902Top 1 Accuracy: 77.76%

Rank counts only results with a code link.