This is an under-develop extension of RapidMiner supported by RapidMiner China. This plug-in integrates RaoidMiner with deep learning features by using the third party library deeplearning4j aka. DL4J, where DL4J is an java-based, open source deeplearning lib developed by Skymind.
- The extension has all it's libraries wrapped inside, thus the only prerequisite to use it is to have RapidMiner 7.0 or above installed.
The current version contains operators implementing thr0ee kinds of neural network models, namely:
Multi-layered Neural Network: This is the general type of neural network. Apply to general labeled numerical example set and could be used for general purpose.
Convolutional Neural Network (CNN): Particularly developed for image process but also works on certain general tasks. Apply to converted labeled image sources, as well as certain general labeled numerical example set.
Word2Vec: For text process only. Works on .txt files which contains raw sentences, currently, only English is supported.