Repository: madlib
Updated Branches:
  refs/heads/master 3db98babe -> 85d09e675


minor docs update to svm and elastic net on cross validation table naming


Project: http://git-wip-us.apache.org/repos/asf/madlib/repo
Commit: http://git-wip-us.apache.org/repos/asf/madlib/commit/85d09e67
Tree: http://git-wip-us.apache.org/repos/asf/madlib/tree/85d09e67
Diff: http://git-wip-us.apache.org/repos/asf/madlib/diff/85d09e67

Branch: refs/heads/master
Commit: 85d09e6750570ca2faf653495f5ee6146c25b536
Parents: 3db98ba
Author: Frank McQuillan <[email protected]>
Authored: Thu Sep 13 12:22:21 2018 -0700
Committer: Frank McQuillan <[email protected]>
Committed: Thu Sep 13 12:22:21 2018 -0700

----------------------------------------------------------------------
 .../modules/elastic_net/elastic_net.sql_in        |  2 ++
 src/ports/postgres/modules/svm/svm.sql_in         | 18 ++++++++++++++++++
 2 files changed, 20 insertions(+)
----------------------------------------------------------------------


http://git-wip-us.apache.org/repos/asf/madlib/blob/85d09e67/src/ports/postgres/modules/elastic_net/elastic_net.sql_in
----------------------------------------------------------------------
diff --git a/src/ports/postgres/modules/elastic_net/elastic_net.sql_in 
b/src/ports/postgres/modules/elastic_net/elastic_net.sql_in
index e30c98c..157851d 100644
--- a/src/ports/postgres/modules/elastic_net/elastic_net.sql_in
+++ b/src/ports/postgres/modules/elastic_net/elastic_net.sql_in
@@ -239,6 +239,8 @@ averaged over all folds and all rows.  For classification, 
the accuracy
 metric used is the ratio of correct classifications.  For regression, the
 accuracy metric used is the negative of mean squared error (negative to
 make it a concave problem, thus selecting \e max means the highest accuracy).
+Cross validation scores are written out to a separate table with the
+user specified name given in the 'validation_result' parameter.
 
 The values of a parameter to cross validate should be provided in a list. For
 example, to regularize with the L1 norm and use a lambda value

http://git-wip-us.apache.org/repos/asf/madlib/blob/85d09e67/src/ports/postgres/modules/svm/svm.sql_in
----------------------------------------------------------------------
diff --git a/src/ports/postgres/modules/svm/svm.sql_in 
b/src/ports/postgres/modules/svm/svm.sql_in
index ccf2c26..ddfa134 100644
--- a/src/ports/postgres/modules/svm/svm.sql_in
+++ b/src/ports/postgres/modules/svm/svm.sql_in
@@ -229,6 +229,24 @@ A summary table named \<model_table\>_summary is also 
created, which has the fol
     </tr>
    </table>
 
+   If cross validation is used, a table is created with a
+   user specified name having the following columns:
+    <table class="output">
+    <tr>
+        <th>...</th>
+        <td>Names of cross validation parameters</td>
+    </tr>
+    <tr>
+        <th>mean_score</th>
+        <td>Mean value of accuracy when predicted on the
+        validation fold, averaged over all folds and all rows.</td>
+    </tr>
+    <tr>
+        <th>std_dev_score</th>
+        <td>Standard deviation of accuracy when predicted on the
+        validation fold, averaged over all folds and all rows.</td>
+    </tr>
+   </table>
 
 @anchor svm_regression
 @par Regression Training Function

Reply via email to