Get started

The Data Graphs Python SDK is a client library that provides comprehensive interaction with the Data Graphs API, including schema management, dataset operations, and bulk data import/export.

Pre-requisites

You will need Python 3.12+ and an api_key for the Data Graphs project you are working with. For write operations you will also need a client_id and client_secret. If you have not generated these already, you can do so in Data Graphs' Applications settings. For more information see Authentication.

Installation

pip install pydatagraphs

Configuration

You can configure the Data Graphs client by passing arguments directly to the constructor.

Read-only access (API key only)

from datagraphs import Client

client = Client(
    project_name="my-project",
    api_key="bZiDs4....."
)

Read and write access (with OAuth credentials)

from datagraphs import Client

client = Client(
    project_name="my-project",
    api_key="bZiDs4.....",
    client_id="sR5dfR.....",
    client_secret="pRM6kR....."
)

Core classes

The SDK provides four main classes:

• Client — Low-level HTTP client for querying, loading, and deleting data
• Gateway — High-level wrapper for deploying projects and bulk import/export
• Schema — In-memory schema builder for defining and manipulating domain models
• Dataset — Dataset representation for organising data within a project

Learn more about the Data Graphs API in the Knowledge Base and the API Reference. The Python SDK repo is available on GitHub.