lightning 2.4.0

The Deep Learning framework to train, deploy, and ship AI products Lightning fast.

Stars: 28159, Watchers: 28159, Forks: 3372, Open Issues: 854

The Lightning-AI/pytorch-lightning repo was created 5 years ago and the last code push was 8 hours ago.
The project is extremely popular with a mindblowing 28159 github stars!

How to Install lightning

You can install lightning using pip

pip install lightning

or add it to a project with poetry

poetry add lightning

Package Details

Author

Lightning AI et al.

License

Apache-2.0

Homepage

https://github.com/Lightning-AI/lightning

PyPi:

https://pypi.org/project/lightning/

Documentation:

https://lightning.ai/lightning-docs

GitHub Repo:

https://github.com/Lightning-AI/lightning

Classifiers

• Scientific/Engineering/Artificial Intelligence
• Scientific/Engineering/Information Analysis

Related Packages

Errors

A list of common lightning errors.

Code Examples

Here are some lightning code examples and snippets.

GitHub Issues

The lightning package has 854 open issues on GitHub

• relax App dependencies for lightning [wip]
• Meta device initialization for FSDP in Trainer
• Cosmic Ray Observatories Data Public
• tests: changed mocked dir tmpdir to tmp_path
• Organize strategy tests folder
• docs: updating logos
• Feature 18367 configurable metric formatting
• WandBLogger: Can't set log_model from LightningCLI due to problem in type hint
• Customizable Metric Formatting for Rich Progress Bar
• Fabric + FSDP + load_raw before setup edge case
• Memory Leak when instantiating Fabric multiple times
• barebones mode should be more forceful
• Update sphinx requirement from <6.0,>5.0 to >5.0,<8.0 in /requirements
• Bump click from 8.1.6 to 8.1.7 in /requirements
• Update fastapi requirement from <0.100.0,>=0.92.0 to >=0.92.0,<0.102.0 in /requirements

See more issues on GitHub