gateplugin-LearningFramework
gateplugin-LearningFramework copied to clipboard
GateNLP
A plugin for the GATE language technology framework for training and using machine learning models. Currently supports Mallet (MaxEnt, NaiveBayes, CRF and others), LibSVM, Scikit-Learn, Weka, and DNN...
This should not be too hard now that we already support python sklearn. The starting point could be to support the models from the tensorflow.contrib.learn package (this was the separate...
We should investigate how GATE and the LF can support sequence to sequence learning.
This got messy when we had to replace the proper weka models with the external approach. Now we do not save the info file if the model is null and...
Think of a way of how to provide most of the functionality to a script that does not import the API. For example, given an initialized training PR, allow to...
This would use the annotation graph convention to represent the ranking problem and support full ranking or just picking the best (by placing the rank into a feature or by...