[
https://issues.apache.org/jira/browse/MAHOUT-1554?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14004448#comment-14004448
]
Karol Grzegorczyk edited comment on MAHOUT-1554 at 5/21/14 7:38 AM:
--------------------------------------------------------------------
I'm submitting a patch ([^statistics.diff]) which introduce weighted average of
precision, recall and F1 score. I decided to compute weighted average, because
it is a default choice in a popular scikit-learn library. Weights are actual
number of test examples (sum of true positives and false negatives).
was (Author: kagrze):
I'm submitting a patch (statistics.diff) which introduce weighted average of
precision, recall and F1 score. I decided to compute weighted average, because
it is a default choice in a popular scikit-learn library. Weights are actual
number of test examples (sum of true positives and false negatives).
> Provide more comprehensive classification statistics
> ----------------------------------------------------
>
> Key: MAHOUT-1554
> URL: https://issues.apache.org/jira/browse/MAHOUT-1554
> Project: Mahout
> Issue Type: Improvement
> Components: Classification
> Reporter: Karol Grzegorczyk
> Priority: Minor
> Fix For: 1.0
>
> Attachments: statistics.diff
>
>
> Currently only limited classification statistics are provided. To better
> understand classification results, it would be worth to provide at lease
> average precision, recall and F1 score.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
