Java -Xmx4g -cp /path/to/ -model /path/to/ -nthreads Ĭurrently, all text is output to standard out.
runs a POS tagger, or any kind of tagging system.Īll of these systems expect plain text files as input, along with a path to a model file.
runs an NER system, or any kind of segmentation system.
There are three classes, one for each kind of system: Command-line UsageĮpic bundles command line interfaces for using parsers, NER systems, and POS taggers (and more generally, segmentation and tagging systems). For instance,Ī named entity recognition system might identify all the people, Segmenters break a sentence into a sequence of fields. Instance, a part-of-speech tagger can identify nouns, verbs, etc. These associate each word in a sentence with a label. Sequence labelers are things like part-of-speech Parsers produce syntactic representations Documentationĭocumentation will (eventually) live at the GitHub wiki: Įpic can be used programmatically or from the command line, usingĮither pretrained models ( see below) or withĬurrently, Epic has support for three kinds of models: parsers, sequence It also includes classes for training high-accuracy syntactic parsers, part-of-speech taggers, name entity recognizers, and more.Įpic is distributed under the Apache License, Version 2.0. EpicĮpic is a structured prediction framework for Scala. I don't have the time to maintain this library, much less modernize it. NLP, like all of AI, has changed a lot since I wrote this back in 2012-2014.