2012/9/4 Alexandre Gramfort <[email protected]>: >> You can just subtract the minimum and divide by the max-min to get >> [0, 1] features and then go from there. >> We could actually add something to do this. > > that would be useful for pipelines so scaling is done on training fold > and not on full data.
Yes, I would be in favor of introducing a new MinMaxScaler class, rename the existing Scaler to UnitVarianceScaler (with backward compat and deprecation warnings) and maybe a RankScaler if someone volunteers to tackle: https://github.com/scikit-learn/scikit-learn/issues/1062 -- 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
