Activation Functions
Hard Swish
Introduced by Howard et al. in Searching for MobileNetV3Hard Swish is a type of activation function based on Swish, but replaces the computationally expensive sigmoid with a piecewise linear analogue:
$$\text{h-swish}\left(x\right) = x\frac{\text{ReLU6}\left(x+3\right)}{6} $$
Source: Searching for MobileNetV3
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 14 | 13.33% |
| Object Detection | 9 | 8.57% |
| Quantization | 6 | 5.71% |
| Classification | 6 | 5.71% |
| Decoder | 6 | 5.71% |
| Semantic Segmentation | 5 | 4.76% |
| Computational Efficiency | 3 | 2.86% |
| Bayesian Optimization | 3 | 2.86% |
| Image Segmentation | 2 | 1.90% |
ReLU6
