[
https://issues.apache.org/jira/browse/FLINK-1732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14581925#comment-14581925
]
ASF GitHub Bot commented on FLINK-1732:
---------------------------------------
Github user shakir774 commented on the pull request:
https://github.com/apache/flink/pull/823#issuecomment-111136272
Given following error while executing the test suite by Travis CI.
Even `import org.apache.flink.ml.common._` is imported, what may cause this
error.
```
[INFO] ^
[ERROR]
/home/travis/build/apache/flink/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/clustering/GMM.scala:95:
error: not found: type Learner
[INFO] class GMM extends Learner[Point,GMModel] {
[INFO] ^
[ERROR]
/home/travis/build/apache/flink/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/clustering/GMM.scala:105:
error: not found: value parameters
[INFO] parameters.add(NumIterations, numIterations)
[INFO] ^
[ERROR]
/home/travis/build/apache/flink/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/clustering/GMM.scala:115:
error: not found: value parameters
[INFO] parameters.add(NumGaussians, numGaussians)
[INFO] ^
[ERROR]
/home/travis/build/apache/flink/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/clustering/GMM.scala:125:
error: not found: value parameters
[INFO] parameters.add(InitialGaussians,initialGaussians)
[INFO] ^
[ERROR]
/home/travis/build/apache/flink/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/clustering/GMM.scala:258:
error: not found: type Transformer
[INFO] case class GMModel(gmm: DataSet[Gaussian]) extends
Transformer[Point, Prediction]
[INFO] ^
[ERROR]
/home/travis/build/apache/flink/flink-staging/flink-ml/src/main/scala/org/apache/flink/ml/clustering/GMM.scala:137:
error: value parameters is not a member of org.apache.flink.ml.clustering.GMM
[INFO] val resultingParameters = this.parameters ++ fitParameters
```
> Add Gaussian mixture model to machine learning library
> ------------------------------------------------------
>
> Key: FLINK-1732
> URL: https://issues.apache.org/jira/browse/FLINK-1732
> Project: Flink
> Issue Type: New Feature
> Components: Machine Learning Library
> Reporter: Till Rohrmann
> Assignee: Alexander Alexandrov
> Labels: ML
>
> For clustering purposes a distribution-based clustering algorithm such as the
> Gaussian mixture model [1] could be helpful.
> Resources:
> [1]
> [http://en.wikipedia.org/wiki/Expectation%E2%80%93maximization_algorithm#Gaussian_mixture]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
