view release on metacpan or  search on metacpan

ExamplesRegression/README_for_regression_tree_examples  view on Meta::CPAN

                README For Regression Tree Examples


In order to become familiar with the regression tree capabilities of this
module, start by executing the following scripts without changing them and
seeing what sort of output is produced by them:

    regression4.pl

    regression5.pl

    regression6.pl

    regression8.pl

IMPORTANT NOTE: With all these scripts, try all three choices for the
                 constructor parameter jacobian_choice.  As mentioned in
                 the main documentation page, it can be set to one of three
                 values: 0, 1, or 2.  Its default is 0.  When set to 0,
                 your regression coefficients are estimated only with
                 linear least-squares and they are NOT refined further with
                 gradient descent.  To benefit from the full power of
                 gradient descent, set jacobian_choice to 2.  However, note
                 that your computation time will go up significantly with a
                 nonzero value for jacobian_choice.



Here is a brief description of what is achieved by each of these four
scripts:

1.  regression4.pl:

    This script involves just one predictor variable and one dependent
    variable. The training data is drawn from the file

          gendata4.csv

    This data file contains strongly nonlinear data.  When you run the
    script regression4.pl, you will see how much better tree regression is
    compared to linear regression.


2.  regression5.pl:

    This script is essentially the same as the previous script except for
    the fact that the training data in the file

          gendata5.csv

    consists of three noisy segments, as opposed to just two in the
    previous case.

         
3.  regression6.pl:

    In this script we deal with the case when we have two predictor
    variables and one dependent variable.  You can think of the data as
    consisting of noisy height values over an (x1,x2) plane.

    The data used in this script is drawn from the csv file

          gen3Ddata1.csv