Rooba
/
lightning
mirror of https://github.com/Lightning-AI/lightning.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

readme: logo 800px (#17108)

This commit is contained in:
Jirka Borovec 2023-03-16 21:56:47 +01:00 committed by GitHub
parent e9937476a2
commit 13e2ddcfba
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 42 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

57
README.md
View File

@ -1,6 +1,6 @@
<div align="center"> <div align="center">
<img alt="Lightning" src="https://pl-public-data.s3.amazonaws.com/assets_lightning/LightningColor.png" width="600px" style="max-width: 100%;"> <img alt="Lightning" src="https://pl-public-data.s3.amazonaws.com/assets_lightning/LightningColor.png" width="800px" style="max-width: 100%;">
<br/> <br/>
<br/> <br/>
@ -9,7 +9,7 @@
**NEW- Lightning 2.0 is featuring a clean and stable API!!** **NEW- Lightning 2.0 is featuring a clean and stable API!!**
---- ______________________________________________________________________
<p align="center"> <p align="center">
<a href="https://www.lightning.ai/">Lightning.ai</a> • <a href="https://www.lightning.ai/">Lightning.ai</a> •
@ -40,7 +40,6 @@
</div> </div>
## Install Lightning ## Install Lightning
Simple installation from PyPI Simple installation from PyPI
@ -92,7 +91,7 @@ pip install -iU https://test.pypi.org/simple/ pytorch-lightning
</details> </details>
<!-- end skipping PyPI description --> <!-- end skipping PyPI description -->
---- ______________________________________________________________________
## Lightning has 3 core packages ## Lightning has 3 core packages
@ -106,7 +105,7 @@ Lightning gives you granular control over how much abstraction you want to add o
<img src="https://pl-public-data.s3.amazonaws.com/assets_lightning/continuum.png" width="80%"> <img src="https://pl-public-data.s3.amazonaws.com/assets_lightning/continuum.png" width="80%">
</div> </div>
---- ______________________________________________________________________
# PyTorch Lightning: Train and Deploy PyTorch at Scale # PyTorch Lightning: Train and Deploy PyTorch at Scale
@ -114,7 +113,7 @@ PyTorch Lightning is just organized PyTorch - Lightning disentangles PyTorch cod
![PT to PL](docs/source-pytorch/_static/images/general/pl_quick_start_full_compressed.gif) ![PT to PL](docs/source-pytorch/_static/images/general/pl_quick_start_full_compressed.gif)
---- ______________________________________________________________________
### Hello simple model ### Hello simple model
@ -130,6 +129,7 @@ import lightning as L
# A LightningModule (nn.Module subclass) defines a full *system* # A LightningModule (nn.Module subclass) defines a full *system*
# (ie: an LLM, difussion model, autoencoder, or simple image classifier). # (ie: an LLM, difussion model, autoencoder, or simple image classifier).
class LitAutoEncoder(L.LightningModule): class LitAutoEncoder(L.LightningModule):
def __init__(self): def __init__(self):
super().__init__() super().__init__()
@ -155,6 +155,7 @@ class LitAutoEncoder(L.LightningModule):
optimizer = torch.optim.Adam(self.parameters(), lr=1e-3) optimizer = torch.optim.Adam(self.parameters(), lr=1e-3)
return optimizer return optimizer
# ------------------- # -------------------
# Step 2: Define data # Step 2: Define data
# ------------------- # -------------------
@ -170,11 +171,13 @@ trainer.fit(autoencoder, data.DataLoader(train), data.DataLoader(val))
``` ```
Run the model on your terminal Run the model on your terminal
``` bash
```bash
pip install torchvision pip install torchvision
python main.py python main.py
``` ```
----
______________________________________________________________________
## Advanced features ## Advanced features
@ -197,6 +200,7 @@ trainer = Trainer(accelerator="gpu", devices=8)
# 256 GPUs # 256 GPUs
trainer = Trainer(accelerator="gpu", devices=8, num_nodes=32) trainer = Trainer(accelerator="gpu", devices=8, num_nodes=32)
``` ```
</details> </details>
<details> <details>
@ -206,6 +210,7 @@ trainer = Trainer(accelerator="gpu", devices=8, num_nodes=32)
# no code changes needed # no code changes needed
trainer = Trainer(accelerator="tpu", devices=8) trainer = Trainer(accelerator="tpu", devices=8)
``` ```
</details> </details>
<details> <details>
@ -246,12 +251,13 @@ trainer = Trainer(logger=loggers.NeptuneLogger())
<details> <details>
<summary>Early Stopping</summary> <summary>Early Stopping</summary>
```python ```python
es = EarlyStopping(monitor="val_loss") es = EarlyStopping(monitor="val_loss")
trainer = Trainer(callbacks=[es]) trainer = Trainer(callbacks=[es])
``` ```
</details> </details>
<details> <details>
@ -261,6 +267,7 @@ trainer = Trainer(callbacks=[es])
checkpointing = ModelCheckpoint(monitor="val_loss") checkpointing = ModelCheckpoint(monitor="val_loss")
trainer = Trainer(callbacks=[checkpointing]) trainer = Trainer(callbacks=[checkpointing])
``` ```
</details> </details>
<details> <details>
@ -271,6 +278,7 @@ trainer = Trainer(callbacks=[checkpointing])
autoencoder = LitAutoEncoder() autoencoder = LitAutoEncoder()
torch.jit.save(autoencoder.to_torchscript(), "model.pt") torch.jit.save(autoencoder.to_torchscript(), "model.pt")
``` ```
</details> </details>
<details> <details>
@ -287,7 +295,7 @@ with tempfile.NamedTemporaryFile(suffix=".onnx", delete=False) as tmpfile:
</details> </details>
---- ______________________________________________________________________
## Advantages over unstructured PyTorch ## Advantages over unstructured PyTorch
@ -300,13 +308,13 @@ with tempfile.NamedTemporaryFile(suffix=".onnx", delete=False) as tmpfile:
- [Tested rigorously with every new PR](https://github.com/Lightning-AI/lightning/tree/master/tests). We test every combination of PyTorch and Python supported versions, every OS, multi GPUs and even TPUs. - [Tested rigorously with every new PR](https://github.com/Lightning-AI/lightning/tree/master/tests). 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). - Minimal running speed overhead (about 300 ms per epoch compared with pure PyTorch).
---- ______________________________________________________________________
<div align="center"> <div align="center">
<a href="https://lightning.ai/docs/pytorch/stable/">Read the PyTorch Lightning docs</a> <a href="https://lightning.ai/docs/pytorch/stable/">Read the PyTorch Lightning docs</a>
</div> </div>
---- ______________________________________________________________________
# Lightning Fabric: Expert control. # Lightning Fabric: Expert control.
@ -314,7 +322,6 @@ Run on any device at any scale with expert-level control over PyTorch training l
Fabric is designed for the most complex models like foundation model scaling, LLMs, diffussion, transformers, reinforcement learning, active learning. Fabric is designed for the most complex models like foundation model scaling, LLMs, diffussion, transformers, reinforcement learning, active learning.
```diff ```diff
+ import lightning as L + import lightning as L
import torch import torch
@ -354,13 +361,13 @@ Fabric is designed for the most complex models like foundation model scaling, LL
- Designed with multi-billion parameter models in mind - Designed with multi-billion parameter models in mind
- Build your own custom Trainer using Fabric primitives for training checkpointing, logging, and more - Build your own custom Trainer using Fabric primitives for training checkpointing, logging, and more
---- ______________________________________________________________________
<div align="center"> <div align="center">
<a href="https://lightning.ai/docs/fabric/stable/">Read the Lightning Fabric docs</a> <a href="https://lightning.ai/docs/fabric/stable/">Read the Lightning Fabric docs</a>
</div> </div>
---- ______________________________________________________________________
# Lightning Apps: Build AI products and ML workflows # Lightning Apps: Build AI products and ML workflows
@ -376,24 +383,28 @@ Lightning Apps remove the cloud infrastructure boilerplate so you can focus on s
# app.py # app.py
import lightning as L import lightning as L
class TrainComponent(L.LightningWork): class TrainComponent(L.LightningWork):
def run(self, x): def run(self, x):
print(f'train a model on {x}') print(f"train a model on {x}")
class AnalyzeComponent(L.LightningWork): class AnalyzeComponent(L.LightningWork):
def run(self, x): def run(self, x):
print(f'analyze model on {x}') print(f"analyze model on {x}")
class WorkflowOrchestrator(L.LightningFlow): class WorkflowOrchestrator(L.LightningFlow):
def __init__(self) -> None: def __init__(self) -> None:
super().__init__() super().__init__()
self.train = TrainComponent(cloud_compute=L.CloudCompute('cpu')) self.train = TrainComponent(cloud_compute=L.CloudCompute("cpu"))
self.analyze = AnalyzeComponent(cloud_compute=L.CloudCompute('gpu')) self.analyze = AnalyzeComponent(cloud_compute=L.CloudCompute("gpu"))
def run(self): def run(self):
self.train.run("CPU machine 1") self.train.run("CPU machine 1")
self.analyze.run("GPU machine 2") self.analyze.run("GPU machine 2")
app = L.LightningApp(WorkflowOrchestrator()) app = L.LightningApp(WorkflowOrchestrator())
``` ```
@ -407,13 +418,13 @@ lightning run app app.py --setup --cloud
lightning run app app.py lightning run app app.py
``` ```
---- ______________________________________________________________________
<div align="center"> <div align="center">
<a href="https://lightning.ai/docs/app/stable/">Read the Lightning Apps docs</a> <a href="https://lightning.ai/docs/app/stable/">Read the Lightning Apps docs</a>
</div> </div>
---- ______________________________________________________________________
## Examples ## Examples
@ -444,7 +455,7 @@ lightning run app app.py
- [Logistic Regression](https://lightning-bolts.readthedocs.io/en/stable/models/classic_ml.html#logistic-regression) - [Logistic Regression](https://lightning-bolts.readthedocs.io/en/stable/models/classic_ml.html#logistic-regression)
- [Linear Regression](https://lightning-bolts.readthedocs.io/en/stable/models/classic_ml.html#linear-regression) - [Linear Regression](https://lightning-bolts.readthedocs.io/en/stable/models/classic_ml.html#linear-regression)
---- ______________________________________________________________________
## Continuous Integration ## Continuous Integration
@ -470,7 +481,7 @@ Lightning is rigorously tested across multiple CPUs, GPUs, TPUs, IPUs, and HPUs
</center> </center>
</details> </details>
---- ______________________________________________________________________
## Community ## Community