Chi-square | Demo | Train in Data

This site uses cookies

Cookies are important for this site to function properly, to guarantee your safety, and to provide you with the best experience. By continuing to browse, you agree to the use of these cookies.
Learn More.

Autoplay
Autocomplete

Previous Lesson Complete and Continue

Feature Selection for Machine Learning

Welcome

Introduction (4:03)
Course curriculum overview (3:33)
Course requirements (3:00)
Course aim (1:44)
How to approach this course
Refer a friend program

Course material

Course material (2:01)
Download Jupyter notebooks
Download datasets
Download presentations
Resources to learn machine learning skills

Feature selection

What is feature selection? (6:15)
Feature selection methods | Overview (6:19)
Filter methods (3:33)
Wrapper methods (5:42)
Embedded methods (3:52)
Moving Forward (4:05)
Open-source packages for feature selection (3:00)

Filter Methods | Basics

Constant, quasi constant, and duplicated features – Intro (4:02)
Constant features (7:53)
Quasi-constant features (7:07)
Duplicated features (5:23)
Install Feature-engine
Drop constant and quasi-constant with Feature-engine (4:20)
Drop duplicates with Feature-engine (5:23)

Filter methods | Correlation

Correlation - Intro (2:41)
Correlation Feature Selection (5:32)
Correlation procedures to select features (3:37)
Correlation | Notebook demo (11:49)
Basic methods plus Correlation pipeline
Correlation with Feature-engine (8:01)
Feature Selection Pipeline with Feature-engine (2:19)
Additional reading resources
Added Treat: A Movie We Recommend 🍿

Filter methods | Statistical measures

Statistical methods – Intro (3:25)
Mutual information (6:11)
Mutual information demo (4:39)
Chi-square test (16:15)
Chi-square | Demo (5:54)
Chi-square considerations (9:19)
Chi2 - calculating the expected frequencies (optional) (3:51)
Chi-square quiz
Anova (5:54)
Anova | Demo (6:10)
Select features based of p-values (10:32)
Basic methods + Correlation + Filter with stats pipeline

Filter Methods | Other methods and metrics

Filter Methods with other metrics (3:04)
Univariate model performance metrics (5:52)
Univariate model performance metrics | Demo (4:23)
KDD 2009: Select features by target mean encoding (6:39)
KDD 2009: Select features by mean encoding | Demo (6:59)
Univariate model performance with Feature-engine (4:54)
Target Mean Encoding Selection with Feature-engine (5:20)
Extra Treat: Our Reading Suggestion 📕

Wrapper methods

Wrapper methods – Intro (6:39)
MLXtend
Step forward feature selection (3:14)
SFS - MLXtend vs Sklearn (4:06)
Step forward feature selection | MLXtend (6:00)
Step forward feature selection | sklearn
Step backward feature selection (3:13)
Step backward feature selection | MLXtend (5:50)
Step backward feature selection | Sklearn
Exhaustive search (2:45)
Exhaustive search | Demo (3:37)

Embedded methods | Linear models

Regression Coefficients – Intro (4:05)
Selection by Logistic Regression Coefficients (6:41)
Selection by Linear Regression Coefficients (2:44)
Coefficients change with penalty (5:26)
Basic methods + Correlation + Embedded method using coefficients

Embedded methods – Lasso regularisation

Regularisation – Intro (5:39)
Lasso (6:39)
A note on SelectFromModel
Basic filter methods + LASSO pipeline

Embedded methods | Trees

Feature Selection by Tree importance | Intro (6:46)
Feature Selection by Tree importance | Demo (3:40)
Feature Selection by Tree importance | Recursively (5:04)
Feature selection with decision trees | review

Hybrid feature selection methods

Introduction to hybrid methods (1:50)
Feature Shuffling - Intro (2:41)
Shuffling features | Demo (8:41)
Recursive feature elimination - Intro (2:21)
Recursive feature elimination | Demo (5:42)
Recursive feature addition - Intro (2:06)
Recursive feature addition | Demo (2:55)
Feature Shuffling with Feature-engine (5:39)
Recursive feature elimination with Feature-engine (4:53)
Recursive feature addition with Feature-engine (3:22)

Congratulations! You did it!

Congratulations
Additional reading resources
Next steps

Chi-square | Demo

Complete and Continue

Discussion

0 comments