data-dumper-weka-package

Weka package that allows listening in on data as it passes through filter pipelines (aka weka.filters.MultiFilter), classifiers or clusterers.

Available schemes

• Dumper schemes

  • weka.core.dump.Null -- dummy, does nothing
  • weka.core.dump.InMemory -- not for GUI use, but API use, as it notifies registered listeners whenever the data changes.
  • weka.core.dump.SaveToDisk -- stores the data in the specified file on disk (auto-detects file format based on extension); allows appending a timestamp to output separate files during cross-validation.

• Integration

  • weka.classifiers.meta.Dumper
  • weka.clusterers.Dumper
  • weka.filters.Dumper

Example usage:

When used in conjunction with the weka.filters.MultiFilter, it is possible to output the data at any given stage in the filter pipeline by inserting weka.filters.Dumper instances where required. Here is an example layout:

- weka.filters.MultiFilter
  |
  + weka.filters.Dumper -dumper "weka.core.dump.SaveToDisk -output-file ./1-initial.arff"
  |
  + weka.filters.unsupervised.attribute.AddNoise"
  |
  + weka.filters.Dumper -dumper "weka.core.dump.SaveToDisk -output-file ./2-with_noise.arff"
  |
  + weka.filters.unsupervised.attribute.Normalize"
  |
  + weka.filters.Dumper -dumper "weka.core.dump.SaveToDisk -output-file ./3-normalized.arff"

Releases

• 2019.9.13

How to use packages

For more information on how to install the package, see:

https://waikato.github.io/weka-wiki/packages/manager/

Maven

Use the following dependency in your pom.xml:

    <dependency>
      <groupId>com.github.fracpete</groupId>
      <artifactId>data-dumper-weka-package</artifactId>
      <version>2019.9.13</version>
      <type>jar</type>
      <exclusions>
        <exclusion>
          <groupId>nz.ac.waikato.cms.weka</groupId>
          <artifactId>weka-dev</artifactId>
        </exclusion>
      </exclusions>
    </dependency>