Magdalena Fuentes

Hey @aahlenst ! Thanks for taking the initiative on this, we're happy to help. Also, thanks @faroit for pointing this out, agree we should discuss on best ways of both...

@PRamoneda and @genisplaja also commented that tests in big datasets are slow (with `test_full_dataset`). They mentioned the idea that we could parallelize them to save some time. @rabitt thoughts?

We discussed offline with @lostanlen that we should write down somewhere the version policy when moving the indexes to Zenodo: - what happens in the future with a new release...

@lostanlen pointed out a fair question, how to include index in zenodo when someone is contributing with a new loader? Are we going to do it ourselves manually? Is it...

Give me a couple of days to check if it's worth starting from a backup version

Closed via #335. Because of size limits of GitHub LFS we decided to move big indexes to zenodo.

I like this idea! What if we make `cleanup` dependent on that boolean, and if we're not checking `checksums` by default when we download we delete the compressed files?

We'll embed the `mir-datasets.yaml` file on the mirdata repo, and add a column that indicates if the dataset has a `mirdata` loader or not. Also as part of the checklist...

This didn't go to far at the end. Thoughts on this moving forward? @rabitt @nkundiushuti

Missing entries now are: Beatport EDM key, cante100, Mridangam Stroke, Saraga. Small inconsistencies: GiantSteps tempo has audio available though in the list it says it doesn't.