Learning Rate Schedules
Step Decay
Step Decay is a learning rate schedule that drops the learning rate by a factor every few epochs, where the number of epochs is a hyperparameter.
Image Credit: Suki Lau
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 38 | 21.71% |
| Object Detection | 37 | 21.14% |
| General Classification | 17 | 9.71% |
| Semantic Segmentation | 10 | 5.71% |
| Classification | 7 | 4.00% |
| Real-Time Object Detection | 6 | 3.43% |
| Instance Segmentation | 5 | 2.86% |
| Object Recognition | 5 | 2.86% |
| Domain Generalization | 4 | 2.29% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
