A plugin for Pentaho's Kettle which allows users to make data forecasts based on Recurrent Neural Network models trained on Weka using the wekaDeeplearning4j package. Code based on the Weka Scoring plugin by Mark Hall.
Switch branches/tags
Nothing to show
Clone or download
Latest commit b07ac75 Aug 26, 2016
Permalink
Failed to load latest commit information.
.idea/libraries add step icon Aug 25, 2016
src/main cleanup Aug 26, 2016
README.md Update README.md Aug 15, 2016
build.sh add step icon Aug 25, 2016
pom.xml add step icon Aug 25, 2016

README.md

rnn-forecaster-plugin

This plugin is a Pentaho Data Integration transformation data mining step which allows users to make predictions on future data using Recurrent Neural Network models created and trained on the Weka Time Series package, using the wekaDeeplearning4j package which allows users to configure RNN models using LSTM layers.