chore: added documentatoin of linear warmup cosine annealing learning rate scheduler

Author: le1nux

docs/components/components.md

@@ -40,6 +40,7 @@
 | scheduler | constant_lr | [ConstantLR](https://pytorch.org/docs/stable/generated/torch.optim.lr_scheduler.ConstantLR.html#torch.optim.lr_scheduler.ConstantLR)| [ConstantLRSchedulerConfig](../../src/modalities/config/config.py) | [LRScheduler](https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate) | Multiplies the learning rate of each parameter group by a small constant factor until the number of steps reaches a pre-defined milestone |
 | scheduler | onecycle_lr | [OneCycleLR](https://pytorch.org/docs/stable/generated/torch.optim.lr_scheduler.OneCycleLR.html#torch.optim.lr_scheduler.OneCycleLR)| [OneCycleLRSchedulerConfig](../../src/modalities/config/config.py) | [LRScheduler](https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate) | Sets the learning rate of each parameter group according to the 1cycle learning rate policy. |
 | scheduler | cosine_annealing_lr | [CosineAnnealingLR](https://pytorch.org/docs/stable/generated/torch.optim.lr_scheduler.CosineAnnealingLR.html#torch.optim.lr_scheduler.CosineAnnealingLR)| [CosineAnnealingLRSchedulerConfig](../../src/modalities/config/config.py) | [LRScheduler](https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate) | Set the learning rate of each parameter group using a cosine annealing schedule |
+| scheduler | linear_warmup_cosine_annealing_lr | [LinearWarmupCosineAnnealingLRScheduler](../../src/modalities/optimizers/lr_schedulers.py) | [LinearWarmupCosineAnnealingLRSchedulerConfig](../../src/modalities/config/config.py) | [LRScheduler](https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate) | Linearly warms up to the base learning rate, then decays with cosine annealing for the remaining training steps |
 
 
 ## Tokenization