From the course: Microsoft Azure Machine Learning Fundamentals
Unlock this course with a free trial
Join today to access over 24,900 courses taught by industry experts.
Lab: Monitor data drift
From the course: Microsoft Azure Machine Learning Fundamentals
Lab: Monitor data drift
With lab Number 4, we are going to look at data drift. Data drift, it's a very important concept. So it's important you know how to do it in real world. So I'm going to show you first, the notebook, what you have to do, and then we'll go through the solution together. So in Lab 4, we are going to see data drift. So first, we need to check if the library for data drift is here with the right version, so we can do that. Then we connect to the workspace as usual. And then the next step is going to create a baseline dataset and the target dataset. So to do that, we are going to use our Titanic datasets as baseline. We're going to register it as a baseline, so you can follow the code here. And then -- so here we do a little trick. We are going to just generate some data. And you see, we changed the age and we change the fare, the price of the ticket, and we are going to create a number of files like that for six weeks of data. We can use the name of the file, or we could use a column in…
Contents
-
-
-
-
-
-
(Locked)
Learning objectives28s
-
(Locked)
Deploy and consume your model7m 46s
-
(Locked)
CI/CD with machine learning6m 41s
-
(Locked)
Monitor data drift and use Application Insights9m 23s
-
(Locked)
Demo: Deploy your model and monitor with Application Insights6m 39s
-
(Locked)
Lab: Monitor data drift3m 9s
-
(Locked)
-
-