TAS-pruned ResNet-50

Reported on 2 benchmarks across 1 task · 1 paper

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

Methodology2 results

Network PruningonImageNet
Accuracy· 2019-05-23
76.2
best: 78.79 (ResNet50-2.3 GFLOPs)
Network Pruning via Transformable Architecture Search arXiv:1905.09717
Network PruningonImageNet
GFLOPs· 2019-05-23
2.3
best: 2.5 (ResNet50 2.5 GFLOPS)
Network Pruning via Transformable Architecture Search arXiv:1905.09717