Hi, Whenever I have to deal with the .arff format, I convert the data into a libsvm format using Weka. The scipy arff loader doesn't handle the sparse arff format.
Then I use sklearn.datasets.load_svmlight_file function or its faster counterpart in https://github.com/mblondel/svmlight-loader. Hope it helps, Arnaud Le 02/03/2013 03:00, Tom Fawcett a écrit : > I’m trying to load up an ARFF (Weka-style .arff) so I can use it with > sklearn’s classifiers. I found scipy’s arff_load function, but getting it > into numpy arrays of data and targets is nontrivial. Is there a facility for > doing this (semi-)automatically? > > I looked through the ./sklearn/datasets directory and I did some google > searches but nothing obvious is uncovered. > > Thanks, > -Tom > > > ------------------------------------------------------------------------------ > Everyone hates slow websites. So do we. > Make your web apps faster with AppDynamics > Download AppDynamics Lite for free today: > http://p.sf.net/sfu/appdyn_d2d_feb > _______________________________________________ > Scikit-learn-general mailing list > Scikit-learn-general@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_feb _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
