OpenML Docs

The OpenML Documentation has grown organically over the years, often using different tools depending on the programming language and type of documentation. Here's an overview of the current status. We are currently working on harmonizing this as well and as soon as possible.

General Concepts

On the OpenML front page

Language: HTML
Source: https://github.com/openml/website/tree/master/openml_OS/views/pages/frontend/guide
Result: https://www.openml.org/guide/bootcamp
Deployment: Admin needs to pull on the server

Wiki page: https://github.com/openml/OpenML/wiki/Basic-Concepts

OpenML Backend

In multiple pages of the WIKI: https://github.com/openml/OpenML/wiki

REST API

Uses Swagger.io to generate a nice web interface that also allows trying out the API calls

Language: YAML/JSON
Source: https://app.swaggerhub.com/apis/openml/openml/1.0.0
Result: https://www.openml.org/api_docs
Deployment: Export from SwaggerHub, push to https://github.com/openml/website/blob/master/downloads/swagger.json

There is currently separate documentation for serving files:

Source: https://app.swaggerhub.com/apis/openml/openml_file/1.0.0
Result: https://www.openml.org/api_data_docs

Python API

Uses Sphinx (http://www.sphinx-doc.org/en/stable/)

Language: reStructuredText
Source: https://github.com/openml/openml-python/tree/master/doc + Docstrings in Python code
Result: https://openml.github.io/openml-python
Deployment: CircleCI job that runs automatically when code is updated

R API

User docs: Uses Vignettes (knitr)

Language: Rmarkdown
Source: https://github.com/openml/openml-r/tree/master/vignettes
Result: https://openml.github.io/openml-r/vignettes/OpenML.html
Deployment: Travis job that (should) run automatically when code is updated

API docs: Uses Roxygen

Language: Rd files
Source: Docstrings in R code
Result: Rd files, viewed as help files in the R shell https://github.com/openml/openml-r/tree/master/man
Deployment: Manually pushed