website

Last update

October 8, 2021

Licence type

Apache-2.0

Supported Languages

Python, SQL

Supported Platforms

Kubernetes
open-source
k8s
online-offline

description

⚠️ New, details subject to change.

Feast is an open-source framework that enables you to access data from your machine learning models. It allows teams to register, ingest, serve, and monitor features in production. Test does not provide a UI or support for feature engineering - it only ingests ready-made features.
Core Features
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store
Governance
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store
Monitoring
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store
User Experience
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store
Feature Computation
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store
Data Ingestion
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store
Feature Storage
Ingestion from Streaming Sources
Ingestion from Batch Sources
Feature Registry/Search
Python
Time Travel
Offline Feature Store
Online Feature Store

Other details

Core Build

Postgres, Redis, Kafka, BigQuery, DynamoDB

Core APIs

Python. Streaming and Batch Ingestion using Spark

Environement

Data Sources

Platform integrations

Architecture