Feature Store - Feature Lookup with Filter

mayank347 — Wed, 13 Jul 2022 13:22:33 GMT

I am working with feature store to save the engineered features. However, for the specific case we have lots of feature table and lot of separate target variables on which we want to train separate models.

Now for each of these model, we can leverage the feature store table but we need to filter out the records based on some condition before joining these feature table to form the final training set. I am wondering if there is a way to include filters while making a feature lookup.

Please find an example below-

Lets say we have some customer data and have saved features into customer demographic, customer education, customer xyz and so on. Now we want to train separate model for different target variable which all can use these engineered features. However, for training model to predict target variable y1 - we need customer demographic data only from states A, B, and C, and join the corresponding features from other feature store tables.

So while performing FeatureLookup can we pass some filter to extract only customer from states A,B, and C

Re: Feature Store - Feature Lookup with Filter

Anonymous — Wed, 10 Aug 2022 17:01:35 GMT

Thanks for taking the time to let us know how to make Databricks even better! @Mayank Srivastava I love that you included a real-life example as well. 😊

I think I know the right PM at Databricks that will be interested in this input. Thanks again for sharing!

topic Feature Store - Feature Lookup with Filter in Machine Learning

Feature Store - Feature Lookup with Filter

Re: Feature Store - Feature Lookup with Filter