Neuroph 2.9 Released
The version 2.9 of framework and GUI tool NeurophStudio has been released!
This release brings some new types of neural networks, image preprocessing tools NeurophStudio, and improved 2D visualization for classification sample in Neuroph Studio.
The most important this release brings Convolutional neural networks to Neuroph. There are also some adapters for other machine learning frameworks Weka nad Java ML which
provides simple integration API.
New features on framework level:
New features in NeurophStudio:
- KMeans Clustering
- Radial Basis Function Network (fully implemented learning in RF layer)
- Convolution Neural Network support
- Adapters for Weka and Java ML libraries
- Image preprocessing tool with set of image preprocessing filters
- Image dataset tool (ease the creation of image variations - scale, rotate, translate images)
- Training wizard - automates the MultiLayerPerceptron training process
- Sample Data Set Generator wizard
- Improved Mutilayer Perceptron Classification Sample
- 2D Visualization of neural network output and weights during the training
Downloads for this release are available at our downloads page