From the course: Machine Learning and AI: Advanced Decision Trees with SPSS
Unlock the full course today
Join today to access over 23,200 courses taught by industry experts.
What is XGBoost?
From the course: Machine Learning and AI: Advanced Decision Trees with SPSS
What is XGBoost?
- [Instructor] Extreme Gradient Boosting. XGBoost has a lot of moving parts, so before we do a demo, let's prepare by mentioning a couple of key points. First, as we'll see, it's implemented through Python. That means you can also read the XGBoost documentation online for the Python library, because it's the same algorithm. Also, loss functions are listed as learning tasks in the SPSS model or interface. We're going to see this. So the second tree in a series of boosted trees is built to figure out what adjustment to the model will move the learning task criteria in the optimal direction. This becomes the target of the tree. This is different than the add a boost style of boosting that we just saw, where the target of the tree remains the same, but the incorrect cases are weighted differently. They're both a kind of boosting, but what is being boosted is different. Now, we're not going to do a deep dive into all of…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
(Locked)
Ensembles4m 48s
-
(Locked)
What is bagging?7m 19s
-
(Locked)
Using bagging for feature selection3m 55s
-
(Locked)
Random forests2m 57s
-
(Locked)
What is boosting?3m 32s
-
(Locked)
What is XGBoost?1m 55s
-
(Locked)
XGBoost Tree node2m 56s
-
(Locked)
Costs and priors5m 11s
-
(Locked)
XGBoost Linear1m 50s
-
(Locked)
-