I believe only http://stackoverflow.com/questions/34167105/using-spark-mls-onehotencoder-on-multiple-columns is currently possible i.e. using multiple stringindexers and then multiple one hot encoders one per column
Madabhattula Rajesh Kumar <mrajaf...@gmail.com> schrieb am Do., 9. Feb. 2017 um 06:38 Uhr: > Hi, > > Do we have a below equivalent preprocessing function in Spark ML? > > from sklearn.preprocessing import MultiLabelBinarizer > > Regards, > > Rajesh > > >