Concept Drift: Monitoring Model Quality in Streaming Machine Learning Applications

preview_player
Показать описание
Most machine learning algorithms are designed to work on stationary data. Yet, real-life streaming data is rarely stationary. Models lose prediction accuracy over time if they are not retrained. Without model quality monitoring, retraining decisions are suboptimal and costly. Here, we review the monitoring methods and evaluate them for applicability in modern fast data and streaming applications.
  1. Akka
  2. Lightbend
  3. fast-data
  4. machine-learning
Рекомендации по теме
Concept Drift: Monitoring 0:53:43

Concept Drift: Monitoring Model Quality in Streaming Machine Learning Applications

Machine Learning Monitoring: 0:05:10

Machine Learning Monitoring: What Is Concept Drift?

Machine Learning Model 0:10:26

Machine Learning Model Drift - Concept Drift & Data Drift in ML - Explanation

ML Drift: Identifying 0:15:25

ML Drift: Identifying Issues Before You Have a Problem

Diagnosing Concept Drift 0:00:30

Diagnosing Concept Drift with Visual Analytics

Monitoring for ML 0:03:05

Monitoring for ML concept drift in a Covid Era

BML21 ID 110 0:07:03

BML21 ID 110 A critical review on concept drift monitoring process ...

Model Monitoring Overview 0:01:51

Model Monitoring Overview - Part 1

2.1. How to 0:05:00

2.1. How to evaluate ML model quality.

Is My Data 0:28:45

Is My Data Drifting? Early Monitoring for Machine Learning Models in Production | PyData Global 2021

Model Drift In 0:08:42

Model Drift In machine learning | Concept drift vs Data Drift in Machine Learning

MLOps Salon: Applying 0:21:19

MLOps Salon: Applying MLOps at Scale -Drift detection on data for monitoring ML models in production

Model Monitoring - 0:06:02

Model Monitoring - Concept and Data Drift - Part 2

What is Concept 0:01:51

What is Concept and Data Drift? | Data Science Fundamentals

Machine Learning Monitoring: 0:03:06

Machine Learning Monitoring: What Is Data Drift?

Data Drift & 0:00:50

Data Drift & Concept Drift in machine learning

Intro to ML 1:48:36

Intro to ML Monitoring: Data Drift, Quality, Bias and Explainability

068-Model Drift 0:03:52

068-Model Drift

Concept Drift Detection 0:44:13

Concept Drift Detection with NannyML | Webinar

Automating Model Monitoring 0:58:33

Automating Model Monitoring and Drift Detection

ML Drift - 0:56:58

ML Drift - How to Identify Issues Before They Become Problems // Amy Hodler // MLOps Meetup #89

Drifting Away: Testing 0:55:08

Drifting Away: Testing ML Models in Production

What is model 0:07:15

What is model monitoring? 🤯 Data Drift? 🧐 Learn the hard-part of data science with FREE tools 🔥 #ml...

SREcon21 - Model 0:12:17

SREcon21 - Model Monitoring: Detecting and Analyzing Data Issues

Комментарии
Автор

Dears, I am looking for an algorithms that can be used to monitor the health of a productive model build based on supervised data .How can I define the time when the model must be updated or retrained with the new available data. The input data of the productive model is changing by nature with time and other factors .

elmugerbi
Автор

Great presentation and explanation Emre.

bandehali
Автор

You could have used a script to compensate for the tough to understand accent you have.

amansinghal