---
id: sagemaker-feature-store-pyspark-3-1
version: "1.1.3"
license: Apache License 2.0
license_treatment: permissive
maintenance: dormant
---

# sagemaker-feature-store-pyspark-3.1 — Amazon SageMaker FeatureStore PySpark Bindings

License: permissive · Maintenance: dormant · Downloads: 378.7K/mo

## What it is and what it does

This package is a Spark connector that bridges Apache Spark DataFrames and Amazon SageMaker FeatureStore. It allows you to ingest feature data directly from Spark into both the online and offline stores of a FeatureGroup, and includes utilities to automatically load feature definitions to streamline FeatureGroup creation.

The connector is designed for teams already running Spark workloads on AWS who want to feed computed features into SageMaker's managed feature store. It requires an existing Spark cluster and SageMaker FeatureStore setup with proper AWS credentials; it is not a standalone tool but rather a bridge between two AWS services.

Use it for:

- Batch ingest computed features from a Spark DataFrame into SageMaker FeatureStore for ML model training.
- Load feature definitions automatically from Spark schema to reduce boilerplate when creating new FeatureGroups.
- Synchronize features computed in a Spark pipeline with SageMaker's online store for real-time inference.
- Populate both online and offline stores in a single operation from a Spark job.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Connects Apache Spark DataFrames to Amazon SageMaker FeatureStore for ingesting feature data into online and offline stores, with automatic feature definition loading.

Yes, if you are already running Spark on AWS and have an active SageMaker FeatureStore deployment. No, if you are looking for a general-purpose feature store or do not have Spark infrastructure in place. Caution: the package is dormant (554 days since last release); verify with AWS documentation that it remains supported for your Spark and SageMaker versions before adopting in production.

## Install

pip install sagemaker-feature-store-pyspark-3-1
uv add sagemaker-feature-store-pyspark-3-1
poetry add sagemaker-feature-store-pyspark-3-1

## Installing sagemaker-feature-store-pyspark-3.1

Before you install: High install friction: requires a pre-built Spark environment and AWS SageMaker setup. Package is dormant (554 days since last release), with no recent maintenance activity visible.

License in practice: Apache License 2.0 is permissive; you can use, modify, and distribute this package freely provided you include the license notice and state significant changes.

Quickstart:

pip install sagemaker-feature-store-pyspark-3-1

from sagemaker_feature_store_pyspark import FeatureStoreManager

# Ingest DataFrame to FeatureGroup online/offline store
manager = FeatureStoreManager()
manager.ingest(spark_dataframe, feature_group_name)

Requires an active Apache Spark environment and AWS SageMaker FeatureStore setup with appropriate IAM credentials configured.

Verify before relying:

- Whether this connector is compatible with Spark versions beyond 3.1 or if it is locked to that version.
- Current maintenance status and whether AWS actively supports this package or recommends alternatives.
- Whether Python version constraints exist despite `requires_python` being unspecified.

## Package facts

- License: Apache License 2.0 (permissive)
- Python support: unspecified
- Install friction: high
- Maintenance: dormant
- Downloads: 378.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

spark sagemaker feature store connector, ingest data to sagemaker featurestore, spark dataframe to aws featurestore, sagemaker feature ingestion spark, aws featurestore pyspark integration, spark connector sagemaker ml features, aws-integration, spark-connector, feature-engineering

[View on SkillFed](https://skillfed.io/packages/sagemaker-feature-store-pyspark-3-1) · [View on PyPI](https://pypi.org/project/sagemaker-feature-store-pyspark-3-1/)