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

Add reference to the `torch.compile` manual (#20066)

This commit is contained in:
awaelchli 2024-07-09 18:18:35 +02:00 committed by GitHub
parent 00cf5c90ca
commit 50af052b31
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 6 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
docs/source-fabric/advanced/compile.rst
View File

@ -272,6 +272,8 @@ As a note, the compilation phase itself will take some time, taking up to severa
For these reasons, we recommend that you don't invest too much time trying to apply ``torch.compile`` during development, and rather evaluate its effectiveness toward the end when you are about to launch long-running, expensive experiments.
Always compare the speed and memory usage of the compiled model against the original model!
For a thorough troubleshooting guide, see `Torch.compile: the missing manual <https://docs.google.com/document/d/1y5CRfMLdwEoF1nTk9q8qEu1mgMUuUtvhklPKJ2emLU8/edit?usp=sharing>`_.
----
@ -313,5 +315,6 @@ Here are a few resources for further reading after you complete this tutorial:
- `GenAI with PyTorch 2.0 blog post series <https://pytorch.org/blog/accelerating-generative-ai-4/>`_
- `Training Production AI Models with PyTorch 2.0 <https://pytorch.org/blog/training-production-ai-models/>`_
- `Empowering Models with Performance: The Art of Generalized Model Transformation Approach <https://pytorch.org/blog/empowering-models-performance/>`_
- `Torch.compile: the missing manual <https://docs.google.com/document/d/1y5CRfMLdwEoF1nTk9q8qEu1mgMUuUtvhklPKJ2emLU8/edit?usp=sharing>`_
|

3
docs/source-pytorch/advanced/compile.rst
View File

@ -241,6 +241,8 @@ As a note, the compilation phase itself will take some time, taking up to severa
For these reasons, we recommend that you don't invest too much time trying to apply ``torch.compile`` during development, and rather evaluate its effectiveness toward the end when you are about to launch long-running, expensive experiments.
Always compare the speed and memory usage of the compiled model against the original model!
For a thorough troubleshooting guide, see `Torch.compile: the missing manual <https://docs.google.com/document/d/1y5CRfMLdwEoF1nTk9q8qEu1mgMUuUtvhklPKJ2emLU8/edit?usp=sharing>`_.
----
@ -282,5 +284,6 @@ Here are a few resources for further reading after you complete this tutorial:
- `GenAI with PyTorch 2.0 blog post series <https://pytorch.org/blog/accelerating-generative-ai-4/>`_
- `Training Production AI Models with PyTorch 2.0 <https://pytorch.org/blog/training-production-ai-models/>`_
- `Empowering Models with Performance: The Art of Generalized Model Transformation Approach <https://pytorch.org/blog/empowering-models-performance/>`_
- `Torch.compile: the missing manual <https://docs.google.com/document/d/1y5CRfMLdwEoF1nTk9q8qEu1mgMUuUtvhklPKJ2emLU8/edit?usp=sharing>`_
|