Build and train PyTorch models and connect them to the ML lifecycle using Lightning App templates, without handling DIY infrastructure, cost management, scaling, and other headaches.
Go to file
Jirka Borovec f9babd1def
ci: build docs once you prepare release (#20507)
2024-12-19 18:02:13 +01:00
.actions Patch jsonargparse for Python >= 3.12.8 (#20479) 2024-12-09 14:35:44 +01:00
.azure Pin setuptools in GPU tests (#20489) 2024-12-10 23:31:31 +01:00
.github ci: build docs once you prepare release (#20507) 2024-12-19 18:02:13 +01:00
_notebooks@1e0e807329 docs: update ref to latest tutorials (#20475) 2024-12-08 12:53:12 +09:00
dockers Pin setuptools in release container (#20488) 2024-12-10 17:55:26 +01:00
docs docs: bump habana (#20506) 2024-12-19 17:56:32 +01:00
examples Code-style changes via pre-commit (#20483) 2024-12-09 17:00:57 +01:00
requirements bump: Torch `2.5` (#20351) 2024-11-12 15:59:08 +01:00
src Prepare Lightning 2.5.0 release (#20505) 2024-12-19 16:58:31 +01:00
tests Add `step` parameter to `TensorBoardLogger.log_hyperparams` (#20176) 2024-12-11 13:09:05 +01:00
.codecov.yml precommit: unify formatting with prettier (#18605) 2023-09-25 14:34:41 +02:00
.git-blame-ignore-revs Add .git-blame-ignore-revs (#16709) 2023-02-10 08:58:03 +00:00
.gitignore Remove the lightning app code (#20039) 2024-07-05 13:56:29 +02:00
.gitmodules Fix repository links (#13304) 2022-06-15 19:33:43 -04:00
.pre-commit-config.yaml bump python 3.9+ (#20413) 2024-11-25 09:20:17 +01:00
.readthedocs.yml Remove the lightning app code (#20039) 2024-07-05 13:56:29 +02:00
CITATION.cff Fix repository links (#13304) 2022-06-15 19:33:43 -04:00
LICENSE update nightly & upgrade Twine (#5458) 2021-01-26 14:29:47 +01:00
Makefile Remove the lightning app code (#20039) 2024-07-05 13:56:29 +02:00
README.md update links (#20431) 2024-11-19 14:10:02 +01:00
SECURITY.md Update email address in SECURITY.md (#17664) 2023-05-21 08:23:50 +02:00
pyproject.toml bump python 3.9+ (#20413) 2024-11-25 09:20:17 +01:00
requirements.txt Remove the lightning app code (#20039) 2024-07-05 13:56:29 +02:00
setup.py bump python 3.9+ (#20413) 2024-11-25 09:20:17 +01:00

README.md

Lightning

The deep learning framework to pretrain, finetune and deploy AI models.

NEW- Deploying models? Check out LitServe, the PyTorch Lightning for model serving


Quick startExamplesPyTorch LightningFabricLightning AICommunityDocs

PyPI - Python Version PyPI Status PyPI - Downloads Conda codecov

Discord GitHub commit activity license

 

Get started

 

Lightning has 2 core packages

PyTorch Lightning: Train and deploy PyTorch at scale.
Lightning Fabric: Expert control.

Lightning gives you granular control over how much abstraction you want to add over PyTorch.

 

Quick start

Install Lightning:

pip install lightning
Advanced install options

Install with optional dependencies

pip install lightning['extra']

Conda

conda install lightning -c conda-forge

Install stable version

Install future release from the source

pip install https://github.com/Lightning-AI/lightning/archive/refs/heads/release/stable.zip -U

Install bleeding-edge

Install nightly from the source (no guarantees)

pip install https://github.com/Lightning-AI/lightning/archive/refs/heads/master.zip -U

or from testing PyPI

pip install -iU https://test.pypi.org/simple/ pytorch-lightning

PyTorch Lightning example

Define the training workflow. Here's a toy example (explore real examples):

# main.py
# ! pip install torchvision
import torch, torch.nn as nn, torch.utils.data as data, torchvision as tv, torch.nn.functional as F
import lightning as L

# --------------------------------
# Step 1: Define a LightningModule
# --------------------------------
# A LightningModule (nn.Module subclass) defines a full *system*
# (ie: an LLM, diffusion model, autoencoder, or simple image classifier).


class LitAutoEncoder(L.LightningModule):
    def __init__(self):
        super().__init__()
        self.encoder = nn.Sequential(nn.Linear(28 * 28, 128), nn.ReLU(), nn.Linear(128, 3))
        self.decoder = nn.Sequential(nn.Linear(3, 128), nn.ReLU(), nn.Linear(128, 28 * 28))

    def forward(self, x):
        # in lightning, forward defines the prediction/inference actions
        embedding = self.encoder(x)
        return embedding

    def training_step(self, batch, batch_idx):
        # training_step defines the train loop. It is independent of forward
        x, _ = batch
        x = x.view(x.size(0), -1)
        z = self.encoder(x)
        x_hat = self.decoder(z)
        loss = F.mse_loss(x_hat, x)
        self.log("train_loss", loss)
        return loss

    def configure_optimizers(self):
        optimizer = torch.optim.Adam(self.parameters(), lr=1e-3)
        return optimizer


# -------------------
# Step 2: Define data
# -------------------
dataset = tv.datasets.MNIST(".", download=True, transform=tv.transforms.ToTensor())
train, val = data.random_split(dataset, [55000, 5000])

# -------------------
# Step 3: Train
# -------------------
autoencoder = LitAutoEncoder()
trainer = L.Trainer()
trainer.fit(autoencoder, data.DataLoader(train), data.DataLoader(val))

Run the model on your terminal

pip install torchvision
python main.py

 

Why PyTorch Lightning?

PyTorch Lightning is just organized PyTorch - Lightning disentangles PyTorch code to decouple the science from the engineering.

PT to PL

 


Examples

Explore various types of training possible with PyTorch Lightning. Pretrain and finetune ANY kind of model to perform ANY task like classification, segmentation, summarization and more:

Task Description Run
Hello world Pretrain - Hello world example Open In Studio
Image classification Finetune - ResNet-34 model to classify images of cars Open In Studio
Image segmentation Finetune - ResNet-50 model to segment images Open In Studio
Object detection Finetune - Faster R-CNN model to detect objects Open In Studio
Text classification Finetune - text classifier (BERT model) Open In Studio
Text summarization Finetune - text summarization (Hugging Face transformer model) Open In Studio
Audio generation Finetune - audio generator (transformer model) Open In Studio
LLM finetuning Finetune - LLM (Meta Llama 3.1 8B) Open In Studio
Image generation Pretrain - Image generator (diffusion model) Open In Studio
Recommendation system Train - recommendation system (factorization and embedding) Open In Studio
Time-series forecasting Train - Time-series forecasting with LSTM Open In Studio

Advanced features

Lightning has over 40+ advanced features designed for professional AI research at scale.

Here are some examples:

Train on 1000s of GPUs without code changes
# 8 GPUs
# no code changes needed
trainer = Trainer(accelerator="gpu", devices=8)

# 256 GPUs
trainer = Trainer(accelerator="gpu", devices=8, num_nodes=32)
Train on other accelerators like TPUs without code changes
# no code changes needed
trainer = Trainer(accelerator="tpu", devices=8)
16-bit precision
# no code changes needed
trainer = Trainer(precision=16)
Experiment managers
from lightning import loggers

# tensorboard
trainer = Trainer(logger=TensorBoardLogger("logs/"))

# weights and biases
trainer = Trainer(logger=loggers.WandbLogger())

# comet
trainer = Trainer(logger=loggers.CometLogger())

# mlflow
trainer = Trainer(logger=loggers.MLFlowLogger())

# neptune
trainer = Trainer(logger=loggers.NeptuneLogger())

# ... and dozens more
Early Stopping
es = EarlyStopping(monitor="val_loss")
trainer = Trainer(callbacks=[es])
Checkpointing
checkpointing = ModelCheckpoint(monitor="val_loss")
trainer = Trainer(callbacks=[checkpointing])
Export to torchscript (JIT) (production use)
# torchscript
autoencoder = LitAutoEncoder()
torch.jit.save(autoencoder.to_torchscript(), "model.pt")
Export to ONNX (production use)
# onnx
with tempfile.NamedTemporaryFile(suffix=".onnx", delete=False) as tmpfile:
    autoencoder = LitAutoEncoder()
    input_sample = torch.randn((1, 64))
    autoencoder.to_onnx(tmpfile.name, input_sample, export_params=True)
    os.path.isfile(tmpfile.name)

Advantages over unstructured PyTorch

  • Models become hardware agnostic
  • Code is clear to read because engineering code is abstracted away
  • Easier to reproduce
  • Make fewer mistakes because lightning handles the tricky engineering
  • Keeps all the flexibility (LightningModules are still PyTorch modules), but removes a ton of boilerplate
  • Lightning has dozens of integrations with popular machine learning tools.
  • Tested rigorously with every new PR. We test every combination of PyTorch and Python supported versions, every OS, multi GPUs and even TPUs.
  • Minimal running speed overhead (about 300 ms per epoch compared with pure PyTorch).


   

Lightning Fabric: Expert control

Run on any device at any scale with expert-level control over PyTorch training loop and scaling strategy. You can even write your own Trainer.

Fabric is designed for the most complex models like foundation model scaling, LLMs, diffusion, transformers, reinforcement learning, active learning. Of any size.

What to change Resulting Fabric Code (copy me!)
+ import lightning as L
  import torch; import torchvision as tv

 dataset = tv.datasets.CIFAR10("data", download=True,
                               train=True,
                               transform=tv.transforms.ToTensor())

+ fabric = L.Fabric()
+ fabric.launch()

  model = tv.models.resnet18()
  optimizer = torch.optim.SGD(model.parameters(), lr=0.001)
- device = "cuda" if torch.cuda.is_available() else "cpu"
- model.to(device)
+ model, optimizer = fabric.setup(model, optimizer)

  dataloader = torch.utils.data.DataLoader(dataset, batch_size=8)
+ dataloader = fabric.setup_dataloaders(dataloader)

  model.train()
  num_epochs = 10
  for epoch in range(num_epochs):
      for batch in dataloader:
          inputs, labels = batch
-         inputs, labels = inputs.to(device), labels.to(device)
          optimizer.zero_grad()
          outputs = model(inputs)
          loss = torch.nn.functional.cross_entropy(outputs, labels)
-         loss.backward()
+         fabric.backward(loss)
          optimizer.step()
          print(loss.data)
import lightning as L
import torch; import torchvision as tv

dataset = tv.datasets.CIFAR10("data", download=True,
                              train=True,
                              transform=tv.transforms.ToTensor())

fabric = L.Fabric()
fabric.launch()

model = tv.models.resnet18()
optimizer = torch.optim.SGD(model.parameters(), lr=0.001)
model, optimizer = fabric.setup(model, optimizer)

dataloader = torch.utils.data.DataLoader(dataset, batch_size=8)
dataloader = fabric.setup_dataloaders(dataloader)

model.train()
num_epochs = 10
for epoch in range(num_epochs):
    for batch in dataloader:
        inputs, labels = batch
        optimizer.zero_grad()
        outputs = model(inputs)
        loss = torch.nn.functional.cross_entropy(outputs, labels)
        fabric.backward(loss)
        optimizer.step()
        print(loss.data)

Key features

Easily switch from running on CPU to GPU (Apple Silicon, CUDA, …), TPU, multi-GPU or even multi-node training
# Use your available hardware
# no code changes needed
fabric = Fabric()

# Run on GPUs (CUDA or MPS)
fabric = Fabric(accelerator="gpu")

# 8 GPUs
fabric = Fabric(accelerator="gpu", devices=8)

# 256 GPUs, multi-node
fabric = Fabric(accelerator="gpu", devices=8, num_nodes=32)

# Run on TPUs
fabric = Fabric(accelerator="tpu")
Use state-of-the-art distributed training strategies (DDP, FSDP, DeepSpeed) and mixed precision out of the box
# Use state-of-the-art distributed training techniques
fabric = Fabric(strategy="ddp")
fabric = Fabric(strategy="deepspeed")
fabric = Fabric(strategy="fsdp")

# Switch the precision
fabric = Fabric(precision="16-mixed")
fabric = Fabric(precision="64")
All the device logic boilerplate is handled for you
  # no more of this!
- model.to(device)
- batch.to(device)
Build your own custom Trainer using Fabric primitives for training checkpointing, logging, and more
import lightning as L


class MyCustomTrainer:
    def __init__(self, accelerator="auto", strategy="auto", devices="auto", precision="32-true"):
        self.fabric = L.Fabric(accelerator=accelerator, strategy=strategy, devices=devices, precision=precision)

    def fit(self, model, optimizer, dataloader, max_epochs):
        self.fabric.launch()

        model, optimizer = self.fabric.setup(model, optimizer)
        dataloader = self.fabric.setup_dataloaders(dataloader)
        model.train()

        for epoch in range(max_epochs):
            for batch in dataloader:
                input, target = batch
                optimizer.zero_grad()
                output = model(input)
                loss = loss_fn(output, target)
                self.fabric.backward(loss)
                optimizer.step()

You can find a more extensive example in our examples



   

Examples

Self-supervised Learning
Convolutional Architectures
Reinforcement Learning
GANs
Classic ML

   

Continuous Integration

Lightning is rigorously tested across multiple CPUs, GPUs and TPUs and against major Python and PyTorch versions.

*Codecov is > 90%+ but build delays may show less
Current build statuses
System / PyTorch ver. 1.13 2.0 2.1
Linux py3.9 [GPUs] Build Status
Linux (multiple Python versions) Test PyTorch Test PyTorch Test PyTorch
OSX (multiple Python versions) Test PyTorch Test PyTorch Test PyTorch
Windows (multiple Python versions) Test PyTorch Test PyTorch Test PyTorch

   

Community

The lightning community is maintained by

  • 10+ core contributors who are all a mix of professional engineers, Research Scientists, and Ph.D. students from top AI labs.
  • 800+ community contributors.

Want to help us build Lightning and reduce boilerplate for thousands of researchers? Learn how to make your first contribution here

Lightning is also part of the PyTorch ecosystem which requires projects to have solid testing, documentation and support.

Asking for help

If you have any questions please:

  1. Read the docs.
  2. Search through existing Discussions, or add a new question
  3. Join our discord.