Dou Shou Qi dataset

[janvanrijn]

We could add the dataset that we created based on this paper: https://arxiv.org/pdf/1604.07312.pdf

I will compile it and upload it to OpenML

[mfeurer]

I'd argue against adding new datasets - we'd have to check whether someone else uploaded new data in the meantime. Also, how big would that dataset be?

[janvanrijn]

I uploaded it to OpenML: https://www.openml.org/d/41001

Approximately 44k instances

[joaquinvanschoren]

Actually, I could upload the mldata datasets to OpenML. They are well-documented and in the ARFF format. I don't know how many fit the benchmark guidelines, but I can do this fairly easily and it would make the benchmark more varied (hopefully many of these are non-trivial).

On Fri, 22 Dec 2017 at 15:43 janvanrijn notifications@github.com wrote:

I uploaded it to OpenML: https://www.openml.org/d/41001

Approximately 44k instances

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/openml/OpenMLFirstBenchmarkSuite/issues/13#issuecomment-353611115, or mute the thread https://github.com/notifications/unsubscribe-auth/ABpQV4yR9m88c4TSHDo89AV37-Nsv3Ngks5tC8AIgaJpZM4RKyTk .

-- Thank you, Joaquin

[mfeurer]

I skimmed the paper and could not find any description of the actual data - did I miss something?

[joaquinvanschoren]

Small update on MLdata: they did a database dump today and will send me the meta-data soon. On Tue, 30 Jan 2018 at 17:56, Matthias Feurer notifications@github.com wrote:

I skimmed the paper and could not find any description of the actual data

• did I miss something?

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/openml/OpenMLFirstBenchmarkSuite/issues/13#issuecomment-361659420, or mute the thread https://github.com/notifications/unsubscribe-auth/ABpQV46elf87O4Liyl4Ght5hNwSI_NQzks5tP0mzgaJpZM4RKyTk .

-- Thank you, Joaquin

[mfeurer]

They're in there now -> close.