| Project Name: | Implementing "How To" section guides and updating DVC 1.0 docs |
| Open source organization: | DVC |
| Mentor: | @jorgeorpinel |
| Technical writer: | @imhardikj |
DVC is a command-line tool like Git but for machine learning projects. I started with DVC for the first time during the project proposal period, by resolving small issues on DVC's GitHub repo. I went through all the existing documentation of DVC and tried to learn how to use DVC.
Next, I started working on the project proposal itself. The project has two major goals. First one was to implement the "How To” section that addresses the common questions related to DVC in detail. The second goal was to update the documentation as per DVC 1.x (dvc.org/blog/dvc-1-0-release) and other relevant updates that were released during the Doc development period.
During the community bonding period, I focussed on resolving issues that were related to updating documentation as per DVC 1.x. I also tried to follow the discussion on GitHub and discord forums so that I can get accustomed using DVC.
At the start of Doc development period, I created a GitHub issue that consisted of a list of DVC command references that had outdated description and examples, and were required to be updated as per 1.x. Working on individual DVC commands involved verifying all the existing examples, adding new ones if required and also improving the description. It helped me in understanding DVC and its working in more detail.
I was also concurrently working on creating a "How-to" section. This section consists of docs that have end-to-end directions on achieving a specific task. Some of these How-to topics are frequently asked across community channels and documenting them in a single section proved to be useful.
The following pull PR included updating move, version, repro, fetch, get, commit, pull, metrics, plots, status, etc.:
-
iterative/dvc.org#1952 (in progress)
-
iterative/dvc.org#1960 (in progress)
-
iterative/dvc.org#1961 (in progress)
How-to docs can be viewed here:
It is the first time that I am contributing to an open-source project, and hence learning Git and contributing using GitHub looked challenging. But I am glad I was able to achieve it. The process also involved learning how to use DVC and understanding how it works. I would like to thank Jorge Orpinel for his support and feedback. DVC being an active community was helpful too.
It has been a positive experience, and I am grateful to have the opportunity to participate in this project. I look forward to continuing my contributions with DVC and also exploring the open-source world.