systemml-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Deron Eriksson (JIRA)" <j...@apache.org>
Subject [jira] [Closed] (SYSTEMML-501) Remove warnings for parameters without default values using MLContext
Date Fri, 16 Sep 2016 01:33:20 GMT

     [ https://issues.apache.org/jira/browse/SYSTEMML-501?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Deron Eriksson closed SYSTEMML-501.
-----------------------------------

> Remove warnings for parameters without default values using MLContext
> ---------------------------------------------------------------------
>
>                 Key: SYSTEMML-501
>                 URL: https://issues.apache.org/jira/browse/SYSTEMML-501
>             Project: SystemML
>          Issue Type: Improvement
>          Components: APIs
>            Reporter: Deron Eriksson
>             Fix For: Not Applicable
>
>
> Given this script:
> {code}
> X = read($Xin)
> m = matrix(nrow(X), rows=1, cols=1)
> n = matrix(ncol(X), rows=1, cols=1)
> write(m, $Mout)
> write(n, $Nout)
> {code}
> If we execute via the following MLContext code:
> {code}
> ml.registerInput("X", sysMlMatrix, numRows, numCols)
> ml.registerOutput("m")
> ml.registerOutput("n")
> val outputs = ml.execute("shape.dml")
> val m = getScalarInt(outputs, "m")
> val n = getScalarInt(outputs, "n")
> {code}
> We see the following warnings:
> {code}
> WARN DMLScript: shape.dml line 6:9 The parameter $Xin either needs to be passed through
commandline or initialized to default value.
> WARN DMLScript: shape.dml line 9:9 The parameter $Mout either needs to be passed through
commandline or initialized to default value.
> WARN DMLScript: shape.dml line 10:9 The parameter $Nout either needs to be passed through
commandline or initialized to default value.
> {code}
> Seeing these warnings can be confusing for an end user since the parameters $Xin, $Mout,
and $Nout aren't needed when using the MLContext API.
> Perhaps Syntactic Validators can be updated to have knowledge that they are being called
from MLContext and not generate these warnings when called from MLContext?



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

Mime
View raw message