The video discusses the background for Generalized Linear Models followed by a coding example using scikit-learn in Python.
Timeline
(Python 3.8)
00:00 - Outline of video
00:44 - Why Generalized Linear Model (GLM)?
04:27 - Components of GLM
05:53 - Link functions
07:08 - Link function
07:39 - Cost function
08:49 - Code snippet
09:35 - Scikit learn docs: The French Motor Third-Party Liability Claims dataset
10:23 - Open Jupyter notebook
10:42 - Data: .fetch_openml()
12:14 - Data: df.describe()
13:02 - Data: create a new feature
14:23 - Data: plot histogram
15:49 - Data: skew
18:20 - Pipeline: log_transformer
19:15 - Pipeline: ColumnTransformer()
25:00 - Data: split train and test
25:54 - DummyRegressor()
28:03 - * * * CORRECTION * * *: changed to "regressor__sample_weight" instead of "..._wight"
28:55 - Create function: score_estimator
33:53 - RidgeRegressor()
36:04 - PoissonRegressor()
37:50 - TweedieRegressor()
39:12 - GammaRegressor()
43:20 - Predictions
52:27 - Ending notes
