📣 Webinar Oct. 17th, 2:00 PM EST | Unraveling prompt engineering

Drift monitoring

Stay on top of data, concept, and model drift, and know when to optimize and retrain your models.

Feature stability

I want to monitor distribution shift for top 5 important features in all fraud models across entire set

Makes sure that your models are reliable and that your top features contribution to ML predictions remains consistent.

  • Track feature importance over time.
  • Track feature importance across datasets.

Training-serving skew

I want to monitor training-serving skew in 4 models across entire set

Detect distribution discrepancies between the data used to train your machine learning model and the data encountered in production.

  • Identify training datasets not representative of your population.
  • Identify feature implementation issues that skewed inferences or scales.

Dataset shift

I want to monitor input drift for all entities in all fraud models across all segments comparing last week to same period last month

Monitor high dimensionality scenarios with dataset shift monitoring to quantify drift for the entire dataset together and drill down to understand precisely what is drifting and where it’s drifting.

  • Detect changes in statistical behaviors.
  • Identify changes in upstream data sources.

Try the community edition

No credit card required. FREE forever.

1
Sign up for Superwise
Easily get started with a free community edition account.

!pip install superwise

2
Build your project

import superwise as sw
project = sw.project(“Fraud detection”)

model = sw.model(project,“Customer a”)

policy = sw.policy(model,drift_template)

3
Start monitoring

Fraud detection

Entire population drift – high probability of concept drift. Open incident investigation →

Fraud detection

Segment “tablet shoppers” drifting.
Split model and retrain.

FREE community edition

Featured resources

Blog

Troubleshooting model drift

There are many types of drift, so how do you troubleshoot model drift before it impacts your business’s bottom line?
Read more
Blog

A hands-on introduction to drift metrics

Instead of focusing on theoretical concepts, this post will explore drift through a hands-on experiment of drift calculations and visualizations. The experiment will help you grasp how the different drift metrics quantify and understand the basic properties of these measures.
Read more
Blog

Concept drift detection basics

This article will illustrate how you can use Layer and Amazon SageMaker to deploy a machine learning model and track it using Superwise.
Read more