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 DNNs through Pytorch and Keras.
We should make it easier to configure the CRF and the way how it gets trained. Either by adding more parameters or by implementing more pre-configured variations or a combination of both.
There are now a bunch of parameters to do this, see the documentation. Should open a new issue for specific additional things that should be made possible.
We should make it easier to configure the CRF and the way how it gets trained. Either by adding more parameters or by implementing more pre-configured variations or a combination of both.