pytorch-lightning
Deep learning framework (PyTorch Lightning / lightning package)
by k-dense-aiFree
Description
pytorch-lightning is an agent skill for the PyTorch Lightning deep-learning framework, which organises PyTorch code into LightningModules and removes boilerplate while keeping full flexibility. It covers configuring Trainers for multiple GPUs and TPUs, automating training workflows, and scaling across devices, along with best practices for neural-network training. The reviewed release is Lightning 2.6.6, including its checkpoint-loading security fixes. The CPU tests used Python 3.11, Torch 2.14.1 and Lightning 2.6.6. Import the lightning package as import lightning as L. The separate pytorch-lightning distribution uses import pytorch_lightning as pl, and the skill warns not to mix the two class namespaces in one application. GPU, TPU, multi-node and online logger examples are source-verified illustrations rather than executed training runs. Use it when you are training a PyTorch model and want the training loop, checkpoints and multi-device setup handled by the framework. It is part of the k-dense-ai scientific agent skills repository on GitHub, which has about 49,600 stars. The listing shows it as free, and it has about 1,900 installs.…
Highlights
- LightningModule organisation of PyTorch code
- Trainer configuration for GPUs and TPUs
- Checkpoint loading with security fixes
- Targets Lightning 2.6.6
- Avoids mixing the two package namespaces
Ratings & Reviews
No ratings yet
Be the first to rate pytorch-lightning.
People also visit
pytorch-lightning FAQ
What does PyTorch Lightning do?
It organises PyTorch training code into LightningModules and handles the training loop and multi-device scaling.
Which version is covered?
Lightning 2.6.6, with checkpoint-loading security fixes, tested on Python 3.11 with Torch 2.14.1.
Which import should I use?
import lightning as L with the lightning package. The pytorch-lightning package uses a different namespace, and the two should not be mixed.
Were GPU and TPU examples run?
No. They are source-verified illustrations, not executed training validation.
How do I install it?
Run npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill pytorch-lightning.
Is it checked for security?
Its security audits show passes from Gen Agent Trust Hub, Socket and Snyk.
What are the best pytorch-lightning alternatives?
Popular alternatives to pytorch-lightning include pytorch-patterns, deepchem and scvi-tools. See all pytorch-lightning alternatives