Datumbox Machine Studying Framework 0.6.1 Launched

The brand new model of Datumbox Machine Studying Framework has been launched! Obtain it now from Github or Maven Central Repository.

What’s new?

The principle focus of model 0.6.1 is to resolve varied bugs, cut back reminiscence consumption and enhance velocity.

Let’s see intimately the modifications of this model:

1. Bug Fixes: A minor subject associated to Unreleased Assets has been detected and stuck on the Checks of Dataset class. Additionally a reminiscence leak was detected and patched on the AutoCloseConnector class.
2. Improved Reminiscence Footprint: The shutdown hooks at the moment are eliminated when shut() known as; this improves reminiscence utilization. Additionally the TypeInference class has been up to date to cut back reminiscence consumption.
3. Pace: The TextClassifier class has been refactored and few velocity enhancements have been launched.
4. Staying Up-to-date: All dependencies and maven plugins used within the challenge have been up to date to the most recent steady variations. Few extra particulars on an important dependencies: The framework now makes use of MapDB 1.0.8 and it’s inside my plans to maneuver to MapDB 2.0 as soon as a steady model is launched. Furthermore I created a Mavenized model of LIBSVM; we at the moment use essentially the most up to date model which is the three.21.

As anticipated the model 0.6.1 is backwards appropriate with the model 0.6.0 of the framework.

Subsequent steps & roadmap

The event of Datumbox Framework will proceed specializing in model 0.7.0. I plan to introduce couple of modifications on the structure of the framework and additional enhance the velocity and reminiscence footprint of the library. For extra details about the roadmap take a look on the earlier launch announcement.

Don’t overlook to obtain the code of Datumbox v0.6.1 from Github. The library is accessible additionally on Maven Central Repository.

I’m wanting ahead to your feedback and suggestions. Pull requests are all the time welcome! 🙂