Objects

Working with structured objects in Clarifeye

Objects in Clarifeye are structured data extracted from your documents using Pydantic models. They enable you to convert unstructured text into queryable, typed data structures.

Overview

Objects allow you to:

• Define custom schemas for data extraction
• Store structured information with type validation
• Link extracted data to source document locations
• Query and analyze structured data at scale

Creating objects

Object structure

Objects are represented as pydantic classes. For example you can create a class like this:

class Person(BaseModel):
    name: str
    age: int
    occupation: str

Creation with the UI

The first way to extract objects from your document is to create visual extractors.

To do so go to the extractors tab > Create a new object extractor

There, you will be able to iterate on your object:

• Pydantic class
• Extraction prompt
• LLM model

Creation with the API

You can also create objects through the API, to do so you will need to leverage the table apis to retrieve data from the blocks table and then write data to the objects table.