I want to read a csv file and normalize the data. If I understand well how hadoop works, the mapper gets the data line by line. I found this formula to normalize : Xnew = (X