Skip to content
Parham's Notes

Deepchecks

Tests for Continuous Validation of ML Models & Data. Deepchecks is a holistic open-source solution for all of your AI & ML validation needs, enabling to thoroughly test your data and models from research to production.

Deepchecks includes:

  • Deepchecks Testing: Running built-in & your own custom Checks and Suites for Tabular, NLP & CV validation (open source).
  • CI & Testing Management: Collaborating over test results and iterating efficiently until model is production-ready and can be deployed (open source & managed offering).
  • Deepchecks Monitoring: Tracking and validating your deployed models behavior when in production (open source & managed offering).

Drift

What Is Distribution Drift?

Distribution drift is a change in the distribution of the data, the label or the relationship between them. Drift happens over time, and can occur either gradually or suddenly.

Common reasons for drift can be natural changes in the data, such as shopping habits changing for an online retailer (for example, during the covid pandemic, online shopping rose significantly and buyer patterns changed), or data acquisition and processing issues, such as camera settings changing accidentally to have lower exposure.