Hopsworks 3.0: The Python-centric Enterprise Feature Store
Feature stores have their roots in data warehouses and data lakes, and much effort in supporting feature engineering for feature stores has been how to make it easier for Data Scientists to write feature logic that will ultimately be executed as SQL. However, Data Scientists' language of choice is Python. Hopsworks 3.0 is the first enterprise feature store to provide Python native support for secure, high performance feature ingestion and serving to enable the Enterprise to build, deploy, and operate machine learning models at scale.
In this talk, we will explore new capabilities in Hopsworks feature store 3.0 and how it can help Data Scientists who love Python to manage their features for training and serving models. We will investigate native Python support for feature engineering, feature pipelines, feature views that represent models in the feature store, transformation functions, and data validation with Great Expectations.
Hopsworks also supports a Python-based domain-specific language for joining and retrieving features from Data Warehouses and Data Lakes, freeing you from having to write SQL to access enterprise data. We will walk through an end-to-end use case in Python showing you how to go from raw data to an operationalized prediction service.
The Feature Store is the essential part of AI infrastructure that helps organisations bring modern enterprise data to analytical and operational ML systems. It is the simplest most powerful way to get your models to production. From anywhere, to anywhere.From months, to minutes.