An AI Engineer technical guide to Feature Store with FEAST

Data teams are starting to realize that operational machine learning requires solving data problems that extend far beyond the creation of data pipelines. Tecton team, highlighted some of the key data challenges that teams face when productionizing ML systems. Accessing the right raw data Building features from raw data Combining features into training data Calculating and serving features in production Monitoring features in production A feature store helps overcoming the above described challenges and FEAST is one of the most popular open source feature store. Feast is an open-source feature store. It is the fastest path to operationalizing analytic data for model training and online inference. Feast enables on-demand transformations to generate features that combine request data with precomputed features (e.g. time_since_last_purchase), with plans to allow light-weight feature engineering. In this technical video you will learn how to get started with FEAST feature store while learning how things works under to hood while you are working with your feature store. The video has the following content: (00:00) Video Start (0:07) Feature Store content intro (2:43) Feature Store - What is it, and how it helps? (4:08)Feature store - Details (7:15) Google Feature Store - Vertex (7:31) DataBricks Feature Store (9:40) Tecton Feature Store - FEAST (11:44) Feature Store Definition (13:07) Jupyter Notebook: Feast Installation/Init (24:19) Understanding Source Data (29:41) Setting Feature Store - Creating registry catalog and online store (33:25) Feast Architecture Review after hands-on example (34:44) Online store (sqlite) review (36:10) Transforming the feature values from source data (38:15) Understanding Online and offline store (41:33) Features added to online store validation (43:05) Machine Learning with online features (43:15) Saving Model (43:48) Using historical data and saved model to score (45:55) Content Review (46:39) GitHub review to Jupyter Notebook (47:13) Plans to use Postgresql in place of sqllite as online store (47:46) Credits Please visit: ------------------ Prodramp LLC https://prodramp.com @prodramp   / prodramp   Content Creator: Avkash Chauhan (@avkashchauhan)   / avkashchauhan   Tags: #ai #aicloud #h2oai #driverlessai #machinelearning #cloud #mlops #featurestore #tecton #aws #databricks #featureengineering #mlmodel #azureml #sagemaker #model #collaboration #h2ohydrogentorch #pytorch #tensorflow #h2oai