Encog 2.4 Released for Java and .Net

Today we released Encog 2.4 for both Java and .Net. This is a very major upgrade to Encog. It contains the following enhancements. Encog is an advanced machine learning framework for Java.

Encog is an advanced neural network and artificial intelligence library. Encog contains classes to create a wide variety of networks, as well as support classes to normalize and process data for these neural networks. Encog trains using multithreaded resilient propagation. A GUI based workbench is also provided to help model and train neural networks. Encog has been in active development since 2008.

Encog is available for Java, .Net and Silverlight.

An article recently compared Encog to some of the other Java Open Source Neural Network frameworks. Encog was reviewed very favorably compared to the other frameworks. You can read the article here.

• Addition of OpenCL Support: you can now use your GPU(s) to help accelerate neural network training.
• Neuroevolution of Augmenting Topologies (NEAT): Exciting new neural network architecture that automatically evolves the hidden structure of a neural network.
• Levenberg Marquadt: An advanced training method that will often converge faster than any of the other perception training methods used by Encog.
• Workbench Enhancments: Workbench now uses a more advanced tabbed interface.
• Encog Cloud: You can now track training progress remotely from our website. This is the foundation upon which we are going to build our grid computing.

For more information visit the Encog homepage. The Encog versions can be downloaded from the following sources.

• Encog Java Downloads
• Encog .Net Downloads