Take a dive into the Superwise model observability platform capabilities.
Everything you need to observe ML system behaviors and keep your ML healthy in production.
Easily create, customize, and automate your ML monitoring with our library of metrics, policies, and notification channels.
Hit the ground running with 100+ pre-built and fully customizable metrics for data, drift, performance, bias, and explainability.
Everything you need to get started with Superwise, from tutorials to recipes and API references.
Need some help getting started with model observability? Our team will walk you through everything you need to know.
Learn how model observability can help you and your team monitor ML.
Whitepapers, use cases, and research. Everything you need effectively assure the health of your models in production.
Leading ML practitioners from across the globe on what it takes to keep ML running smoothly in production.
Everything you need to know about all types of drift including concept drift, data drift, and model drift.
A framework for building, testing, and implementing a robust model monitoring strategy.
Who we are, how we got here, and where we’re going.
What’s new with Superwise.
Join our webinars on ML observability and meet the teams at events across the globe.
Make a Superwise move and join our team.
Need help getting started? Looking to colaborate? Contact us!
Elemeta is an open-source library in Python for metafeature extraction. With it, you will be able to explore, monitor, and extract features from unstructured data through enriched tabular representations. It provides a straightforward Python API for metadata extraction from unstructured data like text and images.
Monitor contextual metafeatures from Elemeta with Superwise to identify changes in the input or output behaviors of your end users and models.
Monitor statistical metafeatures from Elemeta with Superwise to identify changes in the distribution and representation of your NLP use case.
!pip install superwise
import superwise as sw
project = sw.project(“Fraud detection”)
model = sw.model(project,“Customer a”)
policy = sw.policy(model,drift_template)
Entire population drift – high probability of concept drift. Open incident investigation →
Segment “tablet shoppers” drifting. Split model and retrain.