How can we contribute to Docs?

[init27]

Hello Hello!

I found your work via R/ML. Thanks for creating the framework!

I'm not a Time Series expert and neither do I use it at work. However, I noticed that the documentation is quite bare bones at the moment.

I would love to contribute and was curious how can we get started or what might be the best first places to contribute?

Thanks in advance!

[AzulGarza]

Hi @init27! 👋

We use nbdev to develop neuralforecast. You can find the documentation here.

Broadly speaking:

• Once you have made a git clone of the repo, the first thing you need to do is create a conda environment with the environment.yml file.
• After that, you have to activate the environment and install the library interactively using pip install -e ., and run nbdev_install_git_hooks.
• From here, you can modify the documentation and the library. You have to change the notebooks inside the nbs folder (revising the cells in markdown changes to documentation).
• Finally, it remains to reflect the changes in the python modules and html files. For that, just run nbdev_build_lib (changes in modules) and nbdev_build_docs (changes in the documentation). Before doing push, it is also good practice to run nbdev_clean_nbs (to clean notebooks cache), nbdev_diff_nbs (to verify that there are no differences between notebooks and scripts), and finally nbdev_test_nbs to run the notebooks and verify that everything runs fine. (All these steps can be executed using make nbdev_test_flow).
• You can see changes in the documentation locally by running make docs_serve.

Currently, the only file we do not accept modifications to is index.ipynb. :)

For a more detailed explanation you can see our contributing guidelines. 📝

We are excited to see your contributions! 🎉 🔥
Please let me know if you have more questions. :)