Give us a star ⭐️ on GitHub to support the project!
Get real-time visibility, detect issues, and fix them fast. This is a complete solution with a logging module and user interface to stay on top of your production ML models.
Get a live view of all your production ML systems. Track performance, data quality, and drift consistently. Know when it's time to intervene or retrain the models.
Drill down to specific periods, features, or data segments. Understand environment changes to make sure your models never go stale.
Take action once you spot issues. Get to the root cause faster with pre-built summaries and plots. Switch flexibly from monitoring UI to Jupyter notebook.
Share insights with the team and stakeholders. Communicate the value the model brings and how well it works. Boost collaboration and build trust in ML systems.
Capture and track data summaries. Monitor compliance with data quality KPIs over time.
Monitor input and output distributions. Take action when the drift score exceeds limits.
Detect model drift, underperforming segments, and changes in the model target.
Monitor embeddings, unstructured text data, and quality of LLM and NLP-powered systems.
Easily add Evidently to existing workflows, no matter where you deploy.
