I have a dataset that I have created in Python with a category called change_label that has values "buy, hold, sell" (or alternatively I could use -1,0,1).

I am wondering, what is the best way to fit this data so that machine learning algorithms can be applied to it to make predictions?

My options are label encoding, label binzarization, or one-hot encoding. Basically, I thought that simply using sci-kit learn's LabelBinarizer() would be enough. However, I want to be sure. Any advice?

  • 3
    $\begingroup$ @MaxU I have looked through that question and this is not a duplicate as the answer to that question does not answer my specific question. $\endgroup$
    – zsad512
    Sep 10, 2017 at 15:06


Browse other questions tagged or ask your own question.