Metadata-Version: 2.4
Name: lightning-trainer-utils
Version: 0.1.0
Summary: A Python package for using PyTorch Lightning with custom callbacks and model wrappers.
Home-page: https://github.com/manavmahan/lightning-trainer-utils
Author: Manav Mahan Singh
Author-email: manav@genaec.ai
Requires-Python: >=3.12
Description-Content-Type: text/markdown
Dynamic: author
Dynamic: author-email
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: requires-python
Dynamic: summary

# PyTorch Lightning Trainer Utilities

## ML Model Assumptions

### forward

- The batch will be passed as `kwargs` to `forward` function of the ML model.
```python
self.model(**batch)
```

### return
- ML model should return a dict with the following keys:
    - `loss`
    - `loss-dict`
    - `output` [optional]
