Introducing the Encog Workbench 1.0

Encog allows your Java and DotNet applications to create and use neural networks. These neural networks can be saved to Encog data files. These files have the .eg extension. Starting in version 1.1 the Encog workbench is now available. The Encog workbench allows you to create and manipulate these EG files. These files can be saved for later use inside of your Java or DotNet applications.

Seen below is a simple Encog datafile. This shows a neural network and training data.

Viewing an Encog datafile

Any object you see in the Encog datafile can be edited. Below you see training data being edited.

Editing a training set

Neural networks can also be edited. Encog Workbench can edit feedforward, hopfield and SOM type neural networks. Below you see a simple three layer feedforward nueral network being edited.

Editing a neural network

The memory of the neural network is contained in the weight matrix between each of the layers. You can edit the weight matrix between any of the layers. Below you see a weight matrix being edited.

Editing a weight matrix

Sometimes it is more useful to visulaize a neural network than simply look at its numbers. Encog Workbench also allows you to view a neural network graphically. Below you see a simple neural network.

Visualize a neural network

Training can also be performed in Encog Workbench. Encog Workbench can train using backpropagation, simulated annealing or using a genetic algorithm. Below you can see Encog about to train using back propagation.

About to begin training

Statistics are shown as the neural network trains. Below you see the backpropagation training algorithm progressing.

Training with backpropagation

The Encog Workbench can be used with Windows, Macintosh and Linux. If you would like to download the Encog Workbench, use this link.