flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "ASF GitHub Bot (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (FLINK-2157) Create evaluation framework for ML library
Date Wed, 08 Jul 2015 11:55:04 GMT

    [ https://issues.apache.org/jira/browse/FLINK-2157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14618477#comment-14618477
] 

ASF GitHub Bot commented on FLINK-2157:
---------------------------------------

Github user thvasilo commented on a diff in the pull request:

    https://github.com/apache/flink/pull/871#discussion_r34139345
  
    --- Diff: flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/pipeline/Predictor.scala
---
    @@ -233,11 +292,10 @@ trait PredictOperation[Instance, Model, Testing, Prediction] extends
Serializabl
         * @param model The model representation of the prediciton algorithm
         * @return A label for the provided example of type [[Prediction]]
         */
    -  def predict(value: Testing, model: Model):
    -    Prediction
    +  def predict(value: Testing, model: Model): Prediction
     }
     
    -/** Type class for the evaluate operation of [[Predictor]]. This evaluate operation works
on
    +/** Trait for the evaluate operation of [[Predictor]]. This evaluate operation works
on
    --- End diff --
    
    Will change both back.


> Create evaluation framework for ML library
> ------------------------------------------
>
>                 Key: FLINK-2157
>                 URL: https://issues.apache.org/jira/browse/FLINK-2157
>             Project: Flink
>          Issue Type: New Feature
>          Components: Machine Learning Library
>            Reporter: Till Rohrmann
>            Assignee: Theodore Vasiloudis
>              Labels: ML
>             Fix For: 0.10
>
>
> Currently, FlinkML lacks means to evaluate the performance of trained models. It would
be great to add some {{Evaluators}} which can calculate some score based on the information
about true and predicted labels. This could also be used for the cross validation to choose
the right hyper parameters.
> Possible scores could be F score [1], zero-one-loss score, etc.
> Resources
> [1] [http://en.wikipedia.org/wiki/F1_score]



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

Mime
View raw message