Learning Rate Schedules
Linear Warmup With Linear Decay
Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Retrieval | 114 | 12.51% |
| Language Modelling | 102 | 11.20% |
| Question Answering | 58 | 6.37% |
| Large Language Model | 35 | 3.84% |
| Sentiment Analysis | 32 | 3.51% |
| Sentence | 31 | 3.40% |
| Text Classification | 28 | 3.07% |
| Information Retrieval | 21 | 2.31% |
| Text Generation | 19 | 2.09% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
