Datumbox Machine Studying Framework v0.8.1 launched

The Datumbox v0.8.1 has been launched! Obtain it now from Github or Maven Central Repository.

What’s new?

The primary focus of model 0.8.1 is to resolve varied bugs, replace the depedencies and enhance the code structure of the framework. Listed below are the small print:

• Dependencies:
  • Up to date the Maven Compiler, Nexus Staging, Surefire, SLF4J and Logback Basic plugins to the most recent steady variations.
• Code Enhancements & Bug Fixes:
  • FlatDataColletion:
    • The copyCollection2DoubleArray() and copyCollection2Array() strategies have been eliminated.
    • It now implements the Assortment Interface as a substitute of the Iterable.
  • Descriptives:
    • New rely() technique returns the variety of non-null components.
    • All strategies can now deal with null values. Null values are thought of lacking and they’re ignored from the calculations.
  • TextClassifier:
    • The pipeline steps of the Textual content Classifier change to Characteristic Choice, Numerical Standardization and Modeling.
    • The Categorical Encoding step is now not executed because the Textual content Extractor already encodes the phrases as numeric values.

Many due to Jose Luis for his assist in detecting and reproducing a few of the patched bugs. As all the time many thanks my pal and colleague Eleftherios Bampaletakis for his invaluable suggestions.

Don’t overlook to clone the code of Datumbox Framework v0.8.1 from Github, try the Code Examples and obtain the pre-trained Machine Studying fashions from Datumbox Zoo. I’m wanting ahead to your feedback and strategies.