Architecture Search On Cifar 10 Image
評価指標
FLOPS
Params
Percentage error
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | FLOPS | Params | Percentage error |
|---|---|---|---|
| sharpdarts-faster-and-more-accurate | 579M | 3.6M | 1.98 |
| neural-architecture-optimization | - | 144.6M | 1.93 |
| learning-implicitly-recurrent-cnns-through | - | 33.5 | 2.53 |
| neural-architecture-transfer | 291M | 4.6M | 2.1 |
| does-unsupervised-architecture-representation | - | 3.6 | 2.56 |
| a-generic-graph-based-neural-architecture | - | 4.1M | 2.58 |
| path-level-network-transformation-for | - | 14.3M | 2.30 |
| eeea-net-an-early-exit-evolutionary-neural | - | 3.6 | - |
| proxylessnas-direct-neural-architecture | - | 5.7M | 2.08 |
| muxconv-information-multiplexing-in | 200M | 2.1M | 2.0 |
| darts-differentiable-architecture-search | - | 3.4M | 2.83 |
| neural-architecture-transfer | 232M | 4.3M | 2.6 |
| neural-architecture-search-with-reinforcement | - | 27.6M | 2.4 |
| efficient-neural-architecture-search-via-1 | - | 4.6M | 2.89 |
| alphax-exploring-neural-architectures-with-1 | - | 3.59M | - |
| neural-architecture-transfer | 392M | 6.2M | 1.8 |
| neural-architecture-transfer | 468M | 6.9M | 1.6 |
| cate-computation-aware-neural-architecture | - | 4.1 | 2.46 |
| regularized-evolution-for-image-classifier | - | 34.9M | 2.13 |