Training an NER Component with Your Own Data
In this chapter, you will learn how to use your own data to train spaCy’s pre-trained models. We will do that by training a named entity recognition (NER) pipeline, but you can apply the same knowledge to preprocess and train spaCy pipelines for any NLP task. In this chapter, we will focus more on how to collect and label your own data, since we saw how to train models with spaCy’s config.cfg file in Chapter 6.
The learning journey of this chapter includes how to make the best use of Prodigy, the annotation tool from Explosion, and the team behind spaCy. We will also see how to annotate NER data using Jupyter Notebook. After that, we will update the spaCy pipeline’s NER component with this labeled data.
This chapter takes you through a complete machine learning practice, including collecting data, annotating data, and training a model for information extraction.
By the end of this chapter, you’ll be ready...
