The failure is caused due to the randomness of neural network parameter initialization. With very small probability (around 1%), a neural network cannot correctly learn the weights.
2013/7/12 Hudson (JIRA) <[email protected]> > > [ > https://issues.apache.org/jira/browse/HAMA-777?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13707517#comment-13707517] > > Hudson commented on HAMA-777: > ----------------------------- > > FAILURE: Integrated in Hama-Nightly-for-Hadoop-1.x #971 (See [ > https://builds.apache.org/job/Hama-Nightly-for-Hadoop-1.x/971/]) > HAMA-777: FunctionFactory does not work when create IdentityFunction > (yxjiang: rev 1502542) > * /hama/trunk/CHANGES.txt > * /hama/trunk/ml/src/main/java/org/apache/hama/ml/math/FunctionFactory.java > * > /hama/trunk/ml/src/main/java/org/apache/hama/ml/math/IdentityFunction.java > * > /hama/trunk/ml/src/test/java/org/apache/hama/ml/math/TestFunctionFactory.java > > > > FunctionFactory does not work when create IdentityFunction > > ---------------------------------------------------------- > > > > Key: HAMA-777 > > URL: https://issues.apache.org/jira/browse/HAMA-777 > > Project: Hama > > Issue Type: Bug > > Components: math > > Reporter: Yexi Jiang > > Assignee: Yexi Jiang > > Priority: Minor > > Fix For: 0.6.3 > > > > Attachments: HAMA-777.patch, HAMA-777.patch > > > > > > FunctionFactory does not work when create IdentityFunction. It will not > return an instance of IdentityFunction when the input argument is > 'IdentityFunction'. > > Add the implementation of IdentityFunction. > > -- > This message is automatically generated by JIRA. > If you think it was sent incorrectly, please contact your JIRA > administrators > For more information on JIRA, see: http://www.atlassian.com/software/jira > -- ------ Yexi Jiang, ECS 251, [email protected] School of Computer and Information Science, Florida International University Homepage: http://users.cis.fiu.edu/~yjian004/
