Mailing-List: contact issues-help@flink.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@flink.apache.org
Date: Tue, 17 May 2016 12:11:12 +0000 (UTC)
From: "ASF GitHub Bot (JIRA)" <jira@apache.org>
To: issues@flink.apache.org
Message-ID: <JIRA.12827641.1430914098000.208944.1463487072910@Atlassian.JIRA>
In-Reply-To: <JIRA.12827641.1430914098000@Atlassian.JIRA>
References: <JIRA.12827641.1430914098000@Atlassian.JIRA> <JIRA.12827641.1430914098020@arcas>
Subject: [jira] [Commented] (FLINK-1979) Implement Loss Functions
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
archived-at: Tue, 17 May 2016 12:11:14 -0000


    [ https://issues.apache.org/jira/browse/FLINK-1979?page=3Dcom.atlassian=
.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=3D1528=
6480#comment-15286480 ]=20

ASF GitHub Bot commented on FLINK-1979:
---------------------------------------

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

    https://github.com/apache/flink/pull/1985#discussion_r63510204
 =20
    --- Diff: flink-libraries/flink-ml/src/main/scala/org/apache/flink/ml/o=
ptimization/PartialLossFunction.scala ---
    @@ -47,21 +47,106 @@ object SquaredLoss extends PartialLossFunction {
    =20
       /** Calculates the loss depending on the label and the prediction
         *
    -    * @param prediction
    -    * @param label
    -    * @return
    +    * @param prediction The predicted value
    +    * @param label The true value
    +    * @return The loss
         */
       override def loss(prediction: Double, label: Double): Double =3D {
         0.5 * (prediction - label) * (prediction - label)
       }
    =20
       /** Calculates the derivative of the [[PartialLossFunction]]
         *
    -    * @param prediction
    -    * @param label
    -    * @return
    +    * @param prediction The predicted value
    +    * @param label The true value
    +    * @return The derivative of the loss function
         */
       override def derivative(prediction: Double, label: Double): Double =
=3D {
         (prediction - label)
       }
     }
    +
    +/** Logistic loss function which can be used with the [[GenericLossFun=
ction]]
    +  *
    +  *
    +  * The [[LogisticLoss]] function implements `log(1 + exp(prediction*l=
abel))`
    --- End diff --
   =20
    A minus is missing in the exponent of e.


> Implement Loss Functions
> ------------------------
>
>                 Key: FLINK-1979
>                 URL: https://issues.apache.org/jira/browse/FLINK-1979
>             Project: Flink
>          Issue Type: Improvement
>          Components: Machine Learning Library
>            Reporter: Johannes G=C3=BCnther
>            Assignee: Johannes G=C3=BCnther
>            Priority: Minor
>              Labels: ML
>
> For convex optimization problems, optimizer methods like SGD rely on a pl=
uggable implementation of a loss function and its first derivative.


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