Skip to content

Getting Started

Installation

Axetract requires Python 3.12 or higher.

Using pip

pip install axetract

uv add axetract

Basic Usage

The core of Axetract is the AXEPipeline. It orchestrates the flow from raw HTML to structured data.

1. Simple Extraction

Pass a URL and a natural language query:

from axetract import AXEPipeline

pipeline = AXEPipeline.from_config()

result = pipeline.extract(
    "https://news.ycombinator.com",
    query="List the top 5 articles with their titles and points"
)

for article in result.prediction['articles']:
    print(f"{article['title']} ({article['points']} pts)")

2. Using Pydantic Schemas

For production environments, we recommend defining a Pydantic schema to ensure data consistency.

from pydantic import BaseModel
from typing import List
from axetract import AXEPipeline

class Product(BaseModel):
    name: str
    price: float
    availability: bool

pipeline = AXEPipeline.from_config()

result = pipeline.extract(
    "https://example.com/item",
    schema=Product
)

product = Product(**result.prediction)
print(product.name)

Next Steps