Home

Welcome to the Zingg wiki!

Here are some things we are working on/have on the roadmap

• native support for Snowflake using Snowpark. This will let Snowflake users run Zingg without an external Spark cluster
• stop words detection and configuration per column. this will help improve the matching accuracy
• abbreviation detection and matching
• cli for ease of use
• automated unit and e2e testing
• exposing model statistics and convergence so that people learn when to stop
• suggesting or figuring out which fields should be used in matching
• more transformations and experiments to build better models
• geo matching
• unsupervised models

For committers, we want to improve the experience and are working on

• unit and integration testing
• documenting main packages and classes