Neural Architecture Search On Cifar 100 1
평가 지표
FLOPS
PARAMS
Percentage Error
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | FLOPS | PARAMS | Percentage Error |
|---|---|---|---|
| zen-nas-a-zero-shot-nas-for-high-performance | 487M | 2.0M | 15.6 |
| on-constrained-optimization-in-differentiable | - | 3.16M | 17.44 |
| balanced-mixture-of-supernets-for-learning | - | - | - |
| neural-architecture-transfer | 261M | 3.8M | 14.0 |
| muxconv-information-multiplexing-in | 200M | 2.1M | 13.9 |
| m-darts-model-uncertainty-aware | - | 602M | 19.39 |
| du-darts-decreasing-the-uncertainty-of | - | 3.1M | 16.74 |
| b-darts-beta-decay-regularization-for | - | - | 16.52 |
| neural-architecture-transfer | 398M | 6.4M | 12.5 |
| optimizing-neural-architecture-search-using | - | - | 18.83 |
| blockwisely-supervised-neural-architecture | - | - | 11.7 |
| neural-architecture-transfer | 492M | 7.8M | 12.3 |
| neural-architecture-transfer | 796M | 9.0M | 11.7 |