2012/9/7 Lars Buitinck <[email protected]>: > 2012/9/7 Olivier Grisel <[email protected]>: >> Maybe the default feature extraction has changed and made the matrix >> much denser that it used to be for this example? Although recent >> changes to the vectorizer would tend to decrease the number of >> features (min_df=2) hence make the problem smaller to solve. > > What would be the vectorizer settings to make it work? With min_df=1, > max_df=1.0 the ridge classifier still wants all my RAM.
Maybe Vectorizer has changed in other ways between 0.11 and 0.12. Best thing to do would be to vectorize 20 newsgroups with default parameters (as in the example) in each version and compare the `shape` and `data.nbytes` for each resulting coo matrix. -- Olivier http://twitter.com/ogrisel - http://github.com/ogrisel ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
