Transforming Feature Ideas into Machine Learning Inputs
- 1. Transforming Feature Ideas into Machine Learning Inputs Xavier Conort 9th May 2023 - Singapore
- 2. Transforming Feature Ideas into actionable ML inputs 2 FeatureByte is a free and source-available platform designed to empower data enthusiasts who love creating innovative features and improving model accuracy through data. With FeatureByte, you can: ● Quickly create features, ● Seamlessly experiment with features and ● Effortlessly deploy features All without the headache of creating and managing new pipelines.
- 3. Free and source available platform released on May 8! 3 You can access our repository at https://github.com/featurebyte/featurebyte and our documentation at https://docs.featurebyte.com/0.2/. Please check it out and give us your feedback! We're eager to hear your thoughts and improve FeatureByte to better suit your needs.
- 4. Agenda 4 ● Quick introduction to FeatureByte Architecture ● Introduction to the Grocery dataset ● Transform 9 Features Ideas ● Get training data and deploy our new features in production!
- 6. 6 Your DataBricks, Snowflake or Spark data warehouse is used as: ● data source ● compute engine ● feature store The FeatureByte Service, which runs on Docker, acts as the interface between the SDK and your data warehouse. It automatically manages tasks like validating requests, scheduling and executing tasks, and storing metadata. This Python module transpiles the feature logical plan to platform-specific SQL Light-weight installation that leverages your Data Warehouse You interact with the tool through the SDK, which provides an intuitive framework for creating, sharing, experimenting, deploying, and managing features.
- 7. Grocery dataset Note: To access the Grocery dataset or play with FeatureByte's tutorials - you can install a pre-built local Spark data warehouse with pre-populated data, via the command featurebyte.playground().
- 8. 8 We will work with a Catalog where 4 tables were registered from a grocery dataset. Dimension table with static description on products Item table with details on purchased products in customer invoices Event table where each row indicates an invoice Slowly Changing Dimension table that contains data on customers that change over time. A catalog is an object that help organize your feature engineering assets by domain
- 9. Simple observation set to preview features 9 Observation sets are used to materialize historical feature values. They combine entity values and historical points-in-time that you want to learn from. We will work here with a small Pandas DataFrame to preview features when customers are doing a purchase. The observation set is retrieved from a sample of the GROCERYINVOICE table’s view. The column containing points-in-time must be named ‘POINT-IN-TIME’ The column containing entity values must be named with an accepted serving name as defined during the entity registration. View objects are used to prepare data. Work like SQL views but with a syntax similar to Pandas!
- 10. 9 Feature ideas for the Grocery dataset
- 11. 9 feature ideas for the Customer entity 1. Frequency: the count of invoices by Customer in the past 12 weeks 2. Monetary: the amount spent by Customer in the past 12 weeks 3. Recency: the time since the Customer’s last invoice 4. Basket: the sum spent by Customer per Product Group in the past 12 weeks 5. Diversity: the entropy of the Customer’s basket per Product Group in the past 12 weeks 6. Stability: the cosine similarity of the Customer’s basket in the past 2 weeks compared to their basket in the past 12 weeks 7. Similarity: the cosine similarity of the Customer’s basket compared to the baskets of Customers living in the same state in the past 12 weeks 8. Regularity: the entropy of weekdays of the customer invoices in the past 12 weeks 9. Change: whether the Customer’s Address changed in the past 12 weeks 11
- 12. Feature Idea 1: Frequency Count of invoices by Customer in the past 12 weeks 12 The groupby() method enables to group a view by columns representing entities. The aggregate_over() method enables aggregation operation over a window prior to the observation point, here 12 weeks. Feature objects contain the logical plan to compute a feature
- 13. Feature Idea 2: Monetary Amount spent by Customer in the past 12 weeks 13 Aggregation operations include latest, count, sum, average, minimum, maximum, and standard deviation.
- 14. Feature Idea 3: Recency Time since Customer last invoice 14 RequestColumn.point_in_time() enables the creation of features that compare the value of an other feature with the observation point Another example of using this method would be to derive a customer's age from their birth date and the observation point.
- 15. Feature Idea 4: Basket Sum spent by Customer per Product Group in the past 2 and 12 weeks 15 Aggregation can be done across a categorical column, here ProductGroup. When the feature is materialized, a dictionary is returned. In our example, the keys represent the ProductGroup, and their corresponding values represent the total amount spent on each Product Group.
- 16. Feature Idea 5: Diversity Entropy of the Customer’s basket in the past 12 weeks 16 A dictionary feature can be transformed into another feature. We apply here the entropy.
- 17. Feature Idea 6: Stability Cosine Similarity of the Customer’s basket in the past 2 weeks vs their basket in the past 12 weeks 17 2 dictionary features with different windows can be compared using the cosine similarity.
- 18. Feature Idea 7: Similarity Cosine Similarity of the Customer’s basket in the past 12 weeks vs the baskets of Customers living in the same state 18 When 2 dictionary features are grouped with 2 different entities, ● they can be still compared using the cosine similarity. ● the entity of the resulting feature is determined based on the relationship between the 2 entities. In our example, the primary entity of “Customer_Similarity_with_State_12w” is grocerycustomer because grocerycustomer is a child of the frenchstate entity. The feature can be served by providing only the grocerycustomer entity values. Values for frenchstate entity are derived automatically by FeatureByte.
- 19. Feature Idea 8: Regularity Entropy of weekdays of the Customer’s invoices in the past 12 weeks 19 In this example, the keys of the dictionary will represent the weekday, and their corresponding values will represent the total amount spent on each weekday. The weekdays dictionary feature is here transformed into another feature by applying the entropy.
- 20. Feature Idea 9: Change Whether Customer’s Address Changed in the past 12 weeks 20 Change View can be created from a Slowly Changing Dimension (SCD) table to analyze changes that occur in an attribute, here the StreetAddress of the grocerycustomer.
- 21. Get training data and deploy our 9 features in production!
- 22. Audit one feature via its Definition file 22 The feature definition file is critical for maintaining the integrity of a feature. It serves as the single source of truth for the feature, providing an explicit outline of the intended operations of the feature declaration, including inherited operations. The file is automatically generated when a feature is declared or a new version is derived. It is used to generate the final logical execution graph, which is then transpiled into platform-specific SQL for feature materialization.
- 24. Compute training data 24 You can choose to store your training data in the feature store for reuse or audit. If you connect FeatureByte to your data warehouse, the computation of features is performed in the data warehouse, taking advantage of its scalability, stability, and efficiency. In this case, you can work with much larger observation tables with millions of rows.
- 25. Deploy and get shell template for REST API 25
- 26. Thank You! To get started, check out Repo: https://github.com/featurebyte/featurebyte Documentation: https://docs.featurebyte.com/0.2/
- 27. 27 Catalog objects help organize assets by domain View objects are used to prepare data. Work like SQL views but with a syntax similar to Pandas! Table objects centralize essential metadata for feature engineering. 4 types: event table, item table, SCD table and dimension table. Entity objects and relationships facilitate feature organization and serving Feature objects contain the logical plan to compute a feature Python SDK to register, share and manage tables, features and more…