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 larger ones.
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.

Installation

LightKit is available via pip:

pip install lightkit

If you are using Poetry:

poetry add lightkit

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

lightkit-0.5.0.tar.gz (13.0 kB view details)

Uploaded Dec 12, 2022 Source

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

lightkit-0.5.0-py3-none-any.whl (17.2 kB view details)

Uploaded Dec 12, 2022 Python 3

Details for the file lightkit-0.5.0.tar.gz.

Details for the file lightkit-0.5.0-py3-none-any.whl.

Hashes for lightkit-0.5.0-py3-none-any.whl
Algorithm	Hash digest
SHA256	`9df330519e3282435b7fd89af5e3ca831ce7c2b2e3d6a1c33a0c0401a30218ab`
MD5	`5e03c4435a55e5e7304e8abd83bad00a`
BLAKE2b-256	`b5567c2c2b6c77c2237b253beb3b8f73dd8ec4db44a477a10a7acbef9164e4dd`