1. F

    prediction using training data != training data

    (Full disclosure: This is a cross-post from the Kernel Machines Forum) I'm using LIBSVM for regression analysis and, despite double checking my process, I'm not getting sensible results. Specifically, when I apply my training data to my model, the results have significant scaling and offset...
  2. X

    Libsvm: scale data using python

    Libsvm provides a binary for scaling data i.e. svm-scale. How can I scale data to the normalized values using the Python. And then how do I do the scaling for the test samples?