search

openml / openml-r

R package to interface with OpenML
http://openml.github.io/openml-r/
Other
96 stars 37 forks source link

read.arff dislikes some openml datasets #12

Closed jakob-r closed 10 years ago

jakob-r commented 10 years ago

At least these ones do not work:

blacklist = c("baseball","cmc","hypothyroid","mfeat-factors","sick","spambase","mushroom","page-blocks")

May be change foreign::read.arff() to RWeka::read.arff()

dominikkirchhoff commented 10 years ago

Thanks for pointing this out.

Most of these data sets contain special characters in their column names, like "?", ";", " ", ..., so they can't be processed without further manipulation by the user. This must be fixed soon. We're using RWeka::read.arff() now, so data sets with good names shouldn't make trouble anymore. Closing.

berndbischl commented 10 years ago

Should we not leave this open until it is really fixed?

Otherwise reference one where it is alternatively discussed or open a new, dedicated one, please.

dominikkirchhoff commented 10 years ago

Sorry, forgot to reference to this one: https://github.com/openml/OpenML/issues/40