Java

Chapter 4: Using the Encog Workbench

Submitted by jeffheaton on Tue, 04/13/2010 - 18:42

EBooks for Encog C# and Java now on sale

Submitted by jeffheaton on Sun, 02/21/2010 - 10:28

The ebooks for Encog have been released. There is one for Java and another for C#. Both are on sale for $19.99. Paperback versions of each should be out in a few weeks.

Encog 2.3 for Java, C# and Silverlight Released

Submitted by jeffheaton on Sat, 02/20/2010 - 14:10

Java

Encog 2.3 for Java, .Net and Silverlight has been released. This update includes the following:

Scaled Conjugate Gradient (SCG) Training
Greatly Enhanced Self Organizing Map (SOM) Support
Multithreading now on ALL propagation training methods

Also many bug fixes and smaller enhancements that I found necessary as I wrote a book about Encog. More on the book's availability coming soon. To download Encog visit:

http://www.heatonresearch.com/encog

Encog 2.3 for Java Released to Beta

Submitted by jeffheaton on Sat, 01/02/2010 - 08:33

Java

Encog 2.3.0 for Java has been released as beta. All software is complete and javadocs have been updated. The C# version should follow in a few weeks. Encog 2.3 is the version that both books are to be released on. So Encog 2.3 will not go out of beta until both books have been released. At that point Encog 2.4 will begin. I do not expect any of the Encog 2.x releases to “break” the books. When there is an Encog release that will affect the books, and software based on Encog, this will be called Encog 3. Only a few vague plans for Encog 3, at this point.

Encog 2.3(beta) adds Scaled Conjugate Gradient (Levenberg-Marquardt) Training

Submitted by jeffheaton on Mon, 12/21/2009 - 21:03

Java
Java

We just added a new training technique. I really need to test this out more, but so far it outperforms RPROP on anything I've given it. We also have it fully multithreaded, so I really think this is going to become the training algorithm of choice for Encog. It is called Scaled Conjugate Gradient (Levenberg-Marquardt) Training. So far I am really impressed with it! It is quite easy to use.

Train train = new ScaledConjugateGradient(network, trainingSet);

Java only... for now... this will be part of the official 2.3 release which should come out in the Jan/Feb timeframe.

2 comments

Encog 2.2.0 for Java Released

Submitted by jeffheaton on Tue, 11/10/2009 - 20:21

Java

Encog 2.2.0 for Java has been released. This update is much faster and includes extensive multicore support. A DotNet release with similar features is already well underway, and expected in a few weeks. This is an important update for Encog. It includes many bug fixes from 2.1.0, as well as some important new features, such as:

* Considerable performance improvements
* MPROP Training - Similar to RPROP but tuned for MultiCore.
* Normalization - Built in normalization class

Getting Started PDF Released for Encog Java

Submitted by jeffheaton on Sat, 11/07/2009 - 12:02

Java

Encog has been lacking a getting started document for some time. I've completed a 75 page PDF that describes all of the basics of Encog. This is the Java version for now. C# will follow soon. This is a completly free EBook. I will be doing traditional books that go much deeper into Encog soon. You can download it from here. Just scroll down to "Getting Started".

http://www.heatonresearch.com/encog

Applying Multithreading to Resilient Propagation and Backpropagation

Submitted by jeffheaton on Mon, 10/26/2009 - 04:28

This article shows how the Multi Propagation (MPROP) algorithm was implemented for Encog for Java. Though this article focuses on the Java implementation the C# version would be very similar. MPROP is based on resilient propagation, but is designed to work well with multicore computers and gain maximum performance.

Programming Contributions

Submitted by jeffheaton on Sat, 10/24/2009 - 12:52

Are you a Java or C# programmer? Would you like to contribute some time to the Encog project? We are always looking for volunteers, and at all skill levels. You do need to be proficient with Java or C#, but you by no means need to be an AI expert. We are, of course, glad to have AI experts help! But we usually have tasks available that do not require advanced knowledge of AI.