Phil Steitz wrote:
Perhaps it would help if we had overloaded newData methods that accept
different input strategies, but ultimately they will produce a n x m
double array. That way we can provide users with choice.
I was thinking the same thing. The bit that is troubling me is the
omega matrix required by GLS cluttering the OLS interface. Other types
of models (e.g. weighted) will require other data. Could be we need
separate interfaces for the different types of regression, but maybe it
is better to dispense with the abstract interface altogether. The
reason we have interface / implementation separation is to allow
alternative implementations to be plugged in.
Phil - I created a new issue for this refactor:
https://issues.apache.org/jira/browse/MATH-211
For the moment I kept the MultipleLinearRegression interface as common
read-only interface, pushing down the data input to the implementing
classes. IMO there is a benefit in maintaining an interface that
defines what you obtain from regression, regardless of input and
implementation. Also helps with mocking strategies.
The patch attached also incorporates the loadModelData() method that
you had used in the OLS tests - ie it's now been pulled to the abstract
regression class (renamed to newSampleData() for consistency but we can
swap "sample" for "model" - it's just semantics). Tests have been
refactored to use new input method.
Cheers
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
