==============
fairseq2.optim
==============

.. currentmodule:: fairseq2.optim

The optim module provides optimizers and optimization utilities for training
neural networks.

**Coming soon:** This documentation is being developed. The optim module includes:

- AdamW and other optimizers
- Learning rate schedulers
- Gradient clipping utilities
- Optimization configuration

Please refer to the source code and examples in the meantime.