Metadata-Version: 2.1
Name: lightkit
Version: 0.2.0
Summary: Utilities for PyTorch and PyTorch Lightning.
Home-page: https://github.com/borchero/lightkit
License: MIT
Author: Oliver Borchert
Author-email: me@borchero.com
Requires-Python: >=3.8,<3.10
Classifier: Development Status :: 4 - Beta
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Dist: numpy (>=1.20.0,<2.0.0)
Requires-Dist: pytorch-lightning (>=1.5.0,<2.0.0)
Requires-Dist: torch (>=1.8.0,<2.0.0)
Project-URL: Documentation, https://lightkit.borchero.com
Project-URL: Repository, https://github.com/borchero/lightkit
Description-Content-Type: text/markdown

# LightKit

LightKit provides simple utilities for working with PyTorch and PyTorch Lightning. At the moment,
it provides three simple features:

- A data loader for tabular data that is orders of magnitude faster than PyTorch's builtin data
  loader for medium-sized datasets and beyond.
- A mixin for modules that allows to save not only weights, but also the configuration to the file
  system such that it is easier to retrieve trained models.
- A typed base class for estimators that enables users to easily create estimators with PyTorch and
  PyTorch Lightning which are fully compatible with Scikit-learn.

For more details, consult the [documentation](https://lightkit.borchero.com).

## Installation

LightKit is available via `pip`:

```bash
pip install lightkit
```

If you are using [Poetry](https://python-poetry.org/):

```bash
poetry add lightkit
```

