Add crates: forester, openml, arff?

[mbillingr]

Hello there!

I am working on a few machine learning related crates. Now feels like a good time to add them to are-we-learning-yet :) Unfortunately, I am not sure which categories some of them should go into, or if they should be added at all. That's why I did not simply create a pull request. The alleged ML crates are:

• forester: Random Forests and Decision Trees. This one is easy; the "Decision Trees" category feels like a good match.
• openml: An interface to OpenML. It handles data set downloads and running user defined ML algorithms. There does not seem to be a well-matching category¹. "Data Processing" seems like the best fit, but it is more about pre-processing than data set loading and model verification. Any ideas?
• arff deals with the ARFF file format. I'm not sure if this crate should be added at all because it's just a (de)serialization library. Granted, the file format is machine-learning specific (only used by Weka and OpenML as far I know), but does this make it suitable for are-we-learning-yet?

¹_{Maybe it would be appropriate to add a new category, like "Data Sets" or "Evaluation"?}

[xpe]

@mbillingr Hello, some time has passed. Now, it looks like the categories you seek do exist:

• http://www.arewelearningyet.com/decision-trees/
• http://www.arewelearningyet.com/data-preprocessing/

Given that "csv" currently lives in "data processing", I think you can put arff in there as well.

Want to file a PR?

[mbillingr]

Thanks for the update. I'll see if I can get a PR up.