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 | 24.52% |
| Object Detection | 37 | 23.87% |
| General Classification | 17 | 10.97% |
| Semantic Segmentation | 10 | 6.45% |
| Real-Time Object Detection | 6 | 3.87% |
| Instance Segmentation | 5 | 3.23% |
| Object Recognition | 5 | 3.23% |
| Person Re-Identification | 3 | 1.94% |
| Pose Estimation | 2 | 1.29% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
