Simplify Data Preparation for Machine Learning
summary
Amazon SageMaker Feature Store is presented as a way to simplify feature engineering for machine learning by providing a managed service to create, store, and share features for training and inference. The talk covers the problem of repetitive, manual feature creation across projects and shows how SageMaker Feature Store reduces that repetition and streamlines data preparation workflows.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 0:4960% of the time is spent on data preparation
- 3:29and teams all too often start from scratch
- 15:11Feature Store Ingestion API
- 16:24PutRecord Behavior
- 19:55Scalable batch ingestion with PySpark
- 20:43Performance and scale of streaming ingestion
- 24:20Feature retrieval API
- 26:11App-based feature lookup
- 28:08Inference pipeline
- 33:56Inputs and outputs of Lineage
- 34:39Model lineage extends to include feature lineage
related talks
30mAmazon SageMaker Feature Store: Store, discover, & share features for ML apps
33mAccelerate data preparation with Amazon SageMaker Data Wrangler
44mExploratory data analysis and automated feature engineering
58mPrepare data and model features for ML with ease, speed, and accuracy
32mA feature Store for Machine Learning
30mHow Feature Stores Enable Operational ML
39mDevelop your ML project with Amazon SageMaker
35mPanel 03- How to make the Feature Store become a hidden part of the productionalization of ML models