[jira] [Closed] (MADLIB-826) Issue in SVM with svm_gaussian kernel function

["Frank McQuillan (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MADLIB-826?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Frank McQuillan closed MADLIB-826.
----------------------------------
    Resolution: Fixed

Resolved by writing new SVM for scratch for v1.9

Closing this JIRA.

> Issue in SVM with svm_gaussian kernel function
> ----------------------------------------------
>
>                 Key: MADLIB-826
>                 URL: https://issues.apache.org/jira/browse/MADLIB-826
>             Project: Apache MADlib
>          Issue Type: Bug
>            Reporter: Rashmi Raghu
>            Assignee: Rahul Iyer
>             Fix For: v1.9
>
>
> {code:sql}
> -- Using example data from the documentation site
> SELECT madlib.svm_regression('my_train_data',                                                  'myexp10', false, 'madlib.svm_gaussian');
> ERROR:  plpy.SPIError: function madlib.svm_gaussian(double precision[], double precision[]) does not exist  (seg0 slice2 sdw10:40000 pid=4149) (plpython.c:4648)
> CONTEXT:  Traceback (most recent call last):
>   PL/Python function "svm_regression", line 28, in <module>
>     return online_sv.svm_regression( schema_madlib, input_table, model_table, parallel, kernel_func);
>   PL/Python function "svm_regression", line 75, in svm_regression
> PL/Python function "svm_regression"
> raghurdb=# 
> raghurdb=# \df madlib.svm_gaussian
>                                                    List of functions
>  Schema |     Name     | Result data type |                        Argument data types                         |  Type  
> --------+--------------+------------------+--------------------------------------------------------------------+--------
>  madlib | svm_gaussian | double precision | x double precision[], y double precision[], gamma double precision | normal
> (1 row)
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)