First steps

In the following tutorial, we are going to go over a standard classy example, covering Token Classification, specifically Named Entity Recognition, and you are going to learn the basics of how to use classy without (potentially) writing a single line of code.

The Task

Named Entity Recognition is the task of identifying Named Entities in a text linking them to a predefined set of classes, e.g., ORG (organization), LOC (location), PER (person) and O (other) for non-entities. For instance, consider what will be our running example for this whole tutorial:

Barack	Obama	visited	Google	in	California
PER	PER	O	ORG	O	LOC

Our goal is to train a classification model that, given a sequence of tokens as input, outputs a sequence of Named Entity tags corresponding to each token in the sequence.

info

The contents of this tutorial focus on the task of Token Classification, but the general ideas and code structure are very similar for other tasks as well (you can check them out in the documentation).