Analyzing Logistic Regression Performance with ROC Curves [Part 17] | Machine Learning for Beginners

Join Bea Stollnitz, a Principal Cloud Advocate at Microsoft, as she teaches you how to analyze the performance of your logistic regression model using ROC (Receiver Operating Characteristic) curves. We'll be using these to evaluate the Logistic regression classifier built in the previous video using our pumpkin data set 🎃.

What you'll learn:

✅ What a ROC curve is
✅ How a ROC curve helps in evaluating binary classifiers
✅ How a ROC curve relates to a confusion matrix

Bea will guide you through the process of creating an ROC curve using Python in a Juypter Notebook and how to interpret its results to gain insights into your model's performance.

Make sure to subscribe and hit the notification bell 🔔 so you won't miss upcoming videos in the ML for Beginners series!

📙 Follow along:

The Jupyter Notebook to follow along with this lesson is available here:

#Python #ScikitLearn #pandas #LogisticRegression #DataScience #MachineLearning #ml

📚 Learn more:

📇 Connect with Bea:

0:00 - Intro
0:17 - What is an ROC curve?
0:55 - Definition of an ROC curve
1:29 - Choosing a new threshold for logistic regression
2:21 - Plot ROC using multiple classification thresholds
2:43 - Create an ROC curve in code
3:00 - The shape of an ROC curve
3:38 - Reading an ROC curve
4:10 - Calculate the area under the ROC curve

[cerkut]

Great resource for learning / revising the basics. I hope that @MicrosoftDeveloper would not stop here and publishes the videos for all 26 lessons. Looking forward. Thanks.

[martajumi.inranbows]

@Bea_Stollnitz, I wish you made a continuation of this videos with next topics linked with unsupervised learning! This was brilliant and helped me a lot, since videos were so well made and explanations were easy and on point. Thank you.

[machenme]

For those without a foundation, using video introductions is great. Is there a plan for AI for brainers to also produce videos😃