Components
Haystack provides all the tools needed to build a custom NLP application that works for you. This includes everything from prototyping, to annotating data, to deployment.
| Component | Description |
|---|---|
| Documents, Answers and Labels | The core data structures of Haystack. |
| Document Stores | These store your Document corpus and connect to various backend databases. |
| Pipelines | Pipelines allow for easy configuration of Haystack's NLP components. |
| Ready-Made Pipelines | Template Pipelines that cover the most common use cases. |
| REST API | The REST API allows for quick deployment of Pipelines as a service. |
| REST API Definition | Shows the expected input and output of the endpoints in the provided REST API. |
| Annotation Tool | A tool for managing teams of annotators and facilitating the annotation of text for Question Answering and Document Search. |
| Pseudo Label Generator | Use Pseudo Label Generator to create training data for dense retrievers without human annotation. |