search

timeseriesAI / tsai

Time series Timeseries Deep Learning Machine Learning Python Pytorch fastai | State-of-the-art Deep Learning library for Time Series and Sequences in Pytorch / fastai
https://timeseriesai.github.io/tsai/
Apache License 2.0
5.21k stars 651 forks source link

using docments #594

Closed deven-gqc closed 2 years ago

deven-gqc commented 2 years ago

Hello @oguiza, thank you for this amazing library. I wanted to ask you, are you open to using the docments convention for documenting arguments and docstrings?

For example, in SlidingWindow image

I often find myself, looking at the source code to understand the arguments better. I would be happy to create a PR if you are open to this? image

oguiza commented 2 years ago

Hi @deven-gqc, Thanks for your suggestion. I'm very open to it. It'd be great if you submitted a PR for SlidingWindow for example just to agree on a common format before documenting other parts of the library. We'd also need to agree on which functions or classes we want to document this way, as it would take a considerable amount of time to do it for the entire library.

deven-gqc commented 2 years ago

@oguiza I checked the docs again. I saw that a few functions are documented using the docments style. The PR that I've created follows this style of documenting the arguments.

This style was introduced in fastcore a while ago. Here's a reference link. Link

image

image

oguiza commented 2 years ago

Ok, @deven-gqc, I've just merged the PR you have created. Thanks for putting it together. I have started to document new functions with the docments style. But it'd take a lot of time to time to apply it to most functions. The plan is to document some of the most widely used functions, or those with more questions.