A diagnostics and sharing system for ML
Project description
Flor
Flor is a diagnostics and sharing system for the Machine Learning Lifecycle. Flor transparently captures complete execution traces (CETs) and surfaces them as relational tables with minimal human guidance. You can run SQL queries over CETs by annotating your code with log statements --- after your code executes. Run your ML code with confidence: you won't forget to track something important. When you need the value of some expression in your code, just click on it and give it a name, we'll do ETL. Ask your questions in SQL.
Flor is a work in progress. You are welcome to try it now and give us feedback via Github Issues. We're working on making Flor run faster, without sacrificing completeness, transparency, or ease of use. Learn more
License
Flor is licensed under the Apache v2 License.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.