[
https://issues.apache.org/jira/browse/MATH-857?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13451614#comment-13451614
]
Gilles commented on MATH-857:
-----------------------------
bq. I googled this "SVM diff" [...]
That would be (if you work in a Unix-like environment):
{noformat}
svn diff > math857.patch
{noformat}
in order to create a "patch" file that collects the "difference" of your
working copy of the source code
with respect to the version (a.k.a. "revision") of the code on the shared
repository.
The Commons Math project uses the "subversion" software (abbreviated as "svn")
for keeping of the history of the modifications.
Prior to submitting a code for inclusion, you should thus merge it in that
"svn" environment.
You must install the "subversion" software, then "check-out" the Commons Math
code:
{noformat}
svn co https://svn.apache.org/repos/asf/commons/proper/math/trunk
{noformat}
Then you must also install the "maven" software (abbreviated as "mvn"), as it
is used to perform various tasks such as compiling, running the tests,
generating reports (e.g. "CheckStyle").
Once you've checked out the code, the first thing you'd do is go to the "trunk"
directory and run
{noformat}
mvn test
{noformat}
> Include a VIF and TOLERANCE check for a 2 dimensional double array, to
> determine variables that cause multi-colinearity issues and should be
> excluded from the models
> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: MATH-857
> URL: https://issues.apache.org/jira/browse/MATH-857
> Project: Commons Math
> Issue Type: New Feature
> Affects Versions: 3.0
> Environment: can apply to all operating systems
> Reporter: Marios Michaelidis
> Priority: Minor
> Labels: build, test
> Fix For: 3.1
>
> Attachments: FOR TOLERANCE.rar, Multicolinearity.java
>
> Original Estimate: 48h
> Remaining Estimate: 48h
>
> Multicollinearity is a statistical phenomenon in which two or more predictor
> variables in any multiple regression model are highly correlated. Tolerance
> and VIF are checks that allows to avoid optimization failes due to "inability
> to converge". Most of the times, the major packages (SAS, SPSS etc), have a
> check prior to running the model and they exclude variables that might cause
> these kind of problems. It is quite a useful tool to be in common maths.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
