Agnostic learning vs. prior knowledge datasets

Agnostic Learning vs.

Prior Knowledge
Challenge Datasets

Here are the datasets of the challenge. Check the challenge web site.

Download all the AGNOSTIC LEARNING TRACK DATA (45.6 MB)
Download all the PRIOR KNOWLEDGE TRACK DATA (58.9 MB) and the PRIOR KNOWLEDGE (a document describing the data).
or download dataset by dataset:

Name	Domain	Num. ex. (tr/val/te)	Raw data (for the prior knowledge track)	Preprocessed data (for the agnostic learning track)
ADA	Marketing	4147 / 415 / 41471	14 features, comma separated format, 0.6 MB.	48 features, non-sparse format, 0.6 MB.
GINA	Handwriting recognition	3153 / 315 / 31532	784 features, non-sparse format, 7.7 MB.	970 features, non-sparse format, 19.4 MB.
HIVA	Drug discovery	3845 / 384 / 38449	Chemical structure in MDL-SD format, 30.3 MB.	1617 features, non-sparse format, 7.6 MB.
NOVA	Text classif.	1754 / 175 / 17537	Text. 14 MB.	16969 features, sparse format, 2.3 MB.
SYLVA	Ecology	13086 / 1309 / 130857	108 features, non-sparse format, 6.2 MB.	216 features, non-sparse format, 15.6 MB.

Contact:
Isabelle Guyon
Clopinet Enterprises
955, Creston Road,
Berkeley, CA 94708, U.S.A.
Tel/Fax: (510) 524 6211