Kaggle 30 Days of ML (Day 17) - Partial Dependence Plot - Interpretable Machine Learning - XAI

preview_player
Показать описание
This is the Day 17 of Kaggle's 30 Days of ML Challenge where you can learn Machine Learning (based on Python) in 30 days (Kind of). It's not a competition but a challenge to make a habit of coding ML every day. If you haven't registered for the Kaggle Challenge Don't worry, You can follow along my videos every day.

In this Machine Learning Explainability Video, We'll learn about Partial Dependence Plots
✅What is Partial Dependence Plots
✅How does it differ from Feature Importance
✅Where will you use Partial Dependence Plots
✅How to visualize Decision Tree Plot
✅How to build Partial Dependence Plots with PDPBox
✅How to interpret Partial Dependence Plots in English / Business language
✅2D Partial Dependence Plots
✅Wrap up of PDPs

Credit: Kaggle Learn (Dan Becker)
Рекомендации по теме
Комментарии
Автор

Great explanation on the pdp. I am a newbie and I really feel that this video helped me greatly understanding Partial Dependence Plot. I'm highly impressed by the post and of course the presenter. Thank you so much!

bijayamanandhar
Автор

great explenation, but as you said Y axis is CHANGE in target variable (my understanding is that its residual. if it is residual -> higher the goal score higher the residual then its a wrong interpretation right). please clarify this

satishvavilapalli