featureform-enterprise 0.22.5

Package for the Featureform Enterprise Feature Store

Featureform Python Client

Overview

Featureform’s Python client is a SDK for defining, managing and serving resources (e.g. infrastructure providers, data sources, transformations, etc.). At a high level, the API is divided into two parts:

• Registration: register data stores (e.g. PostgreSQL), data sources (e.g. tables or CSV files) as resources or get and/or list previously registered resources
• Serving: retrieve training sets and features for offline training or online inference

Requirements

• Python 3.9-3.12

Setting Up Your Local Development Environment

Step 1: Install gRPC and Protocol Buffer Tooling

See grpc.io for instructions on installing the protocol buffer compiler for your OS and language-specific plugins for Golang (NOTE: the Golang dependencies can also be installed via Homebrew.)

Step 2: Install uv and Sync Dependencies

Install uv if you haven't already. Then sync the project dependencies:

> uv sync --group dev

This creates a virtual environment and installs all required dependencies, including the featureform client in editable mode.

Step 3: Compile API Protocol Buffers and Python Stub

The shell script gen_grpc.sh has been provided for convenience. Change the file access permissions to make it executable and run it:

> chmod +x gen_grpc.sh
> ./gen_grpc.sh

Step 4: Optionally Run Client Test Suite

To ensure your changes haven’t broken the client, run the test suite with the following make target:

> make pytest

Outcome

With the steps above successfully completed, you should have the featureform CLI command accessible via uv run:

> uv run featureform -h

To further verify that your setup is complete and correct, you may optionally walk through the Quickstart tutorial. You may put the definitions.py file at the root of the project, which won’t be ignored by Git, or use a URL to a file (e.g. hosted on GitHub).