lightning-hydra-template
pytorch_tempest
Our great sponsors
lightning-hydra-template | pytorch_tempest | |
---|---|---|
9 | 3 | |
3,645 | 201 | |
- | - | |
5.1 | 5.0 | |
about 1 month ago | 6 months ago | |
Python | Python | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
lightning-hydra-template
- User-friendly PyTorch Lightning and Hydra template for ML experimentation
-
Best practice for saving logits/activation values of model in PyTorch Lightning
I've been trying to learn PyTorch Lightning and Hydra in order to use/create my own custom deep learning template (e.g. like this) as it would greatly help with my research workflow. A lot of the work I do requires me to analyse metrics based on the logits/activations of the model.
-
[D] Is Pytorch Lightning + Wandb a good combination for research?
I can't say for sure whether it is the best combination for research in the long run, but if you do go down that route I have found this template very useful
-
How research scientists structure their code ?
lightning-hydra-template
-
[D] Any research specific PyTorch based boilerplate code?
This lightning + hydra template is quite complete. Great for learning best practices.
-
Typing and testing for torch
A good example is this project template https://github.com/ashleve/lightning-hydra-template. It uses a lot of cool things such as
-
Our template to kickstart your pytorch projects, with list of best practices. Minimal boilerplate code. Leverages Lightning + Hydra. Focused on scalability, reproducibility and fast experimentation.
and many more! (checkout the #Your Superpowers section of the readme)
-
General and feature-rich PyTorch/Hydra project template for rapid and scalable ML experimentation, with a list of best practices
I write a LightningDatamodule. I found it to be an intuitive way to encapsulate any dataset. LightningDatamodule is a simple abstraction providing methods for data download, split, transforms and exposing dataloaders. Would love to see more researchers try out this concept, even in projects which don't use pytorch lightning. Reading LightningDatamodule makes me immedietely see how the dataset is prepared, while it seems like most data science projects throw around data logic across different parts of the pipeline, making it hard to understand what's going on. You can see example of such datamodule here
-
[P] General and feature-rich PyTorch/Hydra template for rapid and scalable ML research/experimentation, with a list of best practices
I feel like most ML people don't use those tools because they simply don't realize all the advantages (especially Hydra seems like a very useful addition to any deep learning project). I focused on structuring the readme in a way, which (I hope) will give you a quick overview - my hope is it can help to spread the word about those frameworks in a broaded community. It incorporates best practices and tricks I gathered over the last couple of months of playing around with it.
pytorch_tempest
- For what reason do you, or don't you, use PyTorch Lightning?
-
If you're using Python, you should probably use a code formatter. (e.g. black)
Here you can see an example of ci in my training pipeline: https://github.com/Erlemar/pytorch_tempest/blob/master/.github/workflows/ci.yml
-
Nicest and cleanest Deep Learning codebases out there
DL is all experimentation. You need a framework that makes modification easy, trackable and reproducible. The closest to achieving that these days is based on pytorch lightning with hydra. Check out this video and repo: https://www.youtube.com/watch?v=w10WrRA-6uI https://github.com/Erlemar/pytorch_tempest
What are some alternatives?
lightning-hydra-template - Deep Learning project template best practices with Pytorch Lightning, Hydra, Tensorboard.
deepchem - Democratizing Deep-Learning for Drug Discovery, Quantum Chemistry, Materials Science and Biology
neptune-client - 📘 The MLOps stack component for experiment tracking
omegaconf - Flexible Python configuration system. The last one you will ever need.
lightning-transformers - Flexible components pairing 🤗 Transformers with :zap: Pytorch Lightning
nn-template - Generic template to bootstrap your PyTorch project.
traingenerator - 🧙 A web app to generate template code for machine learning
pytorch-lightning - Build high-performance AI models with PyTorch Lightning (organized PyTorch). Deploy models with Lightning Apps (organized Python to build end-to-end ML systems). [Moved to: https://github.com/Lightning-AI/lightning]
neptune-contrib - This library is a location of the LegacyLogger for PyTorch Lightning.
black - The uncompromising Python code formatter
garage - A toolkit for reproducible reinforcement learning research.