MDX track: UMX-L and X-UMXL baselines available

Hello all,

just to let you know that you can now find two baselines in the MDX starter-kit :slight_smile:

Please note that both models can not be used for Leaderboard A and B as they used other data for training.

Kind regards

Stefan

1 Like

Hi @StefanUhlich,

Thanks for the baselines!

I just wanted to know about the scale of external data used for UMX-L. It just says private compressed stems, but could you tell us how much extra data it was trained on?

For example, for the XUMX-L baseline, the Zenodo page says 100 hours of private data.

Thanks!

I’m confused.
The rule stated " Systems that are trained on any other data (e.g., also using the test part tt of the DnR dataset) are eligible for Leaderboard B. "

So doesn’t it mean it can be used for Leaderboard B ?
I may have other data in my current training :frowning:

These baselines are for the MDX track. You cannot use extra data for Leaderboards A and B for the MDX track.

For the CDX track, you can use extra data for Leaderboard B.

1 Like

@subatomicseer Thank you for the reply. I was not careful in distinguishing the two tracks, I’m sorry about it.

Hi @subatomicseer

I just wanted to know about the scale of external data used for UMX-L. It just says private compressed stems, but could you tell us how much extra data it was trained on?

I also don’t know the details - only what @faroit wrote here: https://twitter.com/faroit/status/1413409137032060929

@faroit Do you remember how many hours the dataset corresponded to?

Kind regards

Stefan

@subatomicseer umx-l was trained on about 400 * 4 stems hours of stem data.

1 Like

Hello all,

quick update: we renamed XUMX-L to XUMX-M in the starter-kit in order to avoid any confusion with UMX-L and XUMX-L as they were trained on different datasets.

M stands for medium as XUMX-M was trained on ~100 hours of music compared to the ~400 hours of UMX-L.

Kind regards

Stefan

1 Like