This project is a community effort, and everyone is welcome to contribute !
If you are interested in contributing to PyTorch-Ignite's examples, there are many ways to help out. Your contributions may fall into the following categories:
-
It helps us very much if you could report issues you’re facing with:
- Executing the Juypter notebooks, scripts or rendering assets.
- Understanding the language written in any of the notebooks that can be simplified.
-
You would like to add more examples. These fall into two categories:
- Tutorials: Something general which has emphasis on explanation and has self-contained end-to-end code which showcases an Ignite concept or concepts. These are meant for learning purposes when exploring the library. See cifar10 tutorial on distributed training: https://pytorch-ignite.ai/tutorials/cifar10-distributed/ for an example.
- How-to guides: These are very specific and more code-based. They are used to answer a specific question like how to use FastaiLR finder with Ignite or how to do cross validation. Comparing it with the
cifar10tutorial above, if we were to make a how to guide it could go like: how to train a model using multiple gpus with ignite.
Please refer to README.md on how to generate Jupyter notebooks with built-in frontmatter.