Applied Machine Learning 2019 - Lecture 12 - Model Interpretration and Feature Selection

preview_player
Показать описание
Feature importance measures, partial dependence plots.
Univariate and multivariate feature selection, recursive feature selection.

Slides and more materials are on the class website:
  1. Andreas Mueller
  2. machine learning
  3. feature selection
  4. SHAP
  5. LIME
  6. feature importance
Рекомендации по теме
Applied Machine Learning 1:23:21

Applied Machine Learning 2019 - Lecture 01 - Introduction to Machine Learning

Applied Machine Learning 1:22:07

Applied Machine Learning 2019 - Lecture 13 - Parameter Selection and Automatic Machine Learning

[VDT19] Applied machine 0:44:41

[VDT19] Applied machine learning: a few lessons I learned the hard way by Alessandro Giusti

Applied Machine Learning 1:15:23

Applied Machine Learning 2019 - Lecture 10 - Model Evaluation

Applied Machine Learning 1:18:07

Applied Machine Learning 2019 - Lecture 24 - Recap and summary

Applied Machine Learning 1:06:42

Applied Machine Learning 2019 - Lecture 20 - Neural Networks

Applied Machine Learning 1:14:41

Applied Machine Learning 2019 - Lecture 02 - Version control and Testing

Vida Williams - 0:31:13

Vida Williams - 2019 Tom Tom Applied Machine Learning Conference Keynote

Applied Machine Learning 1:16:45

Applied Machine Learning 2019 - Lecture 17 - Introduction to text data

Applied Machine Learning 1:18:24

Applied Machine Learning 2019 - Lecture 18 - Topic Models

AMLD2019 - Christopher 0:33:06

AMLD2019 - Christopher Bishop, Microsoft: Applied Machine Learning: The Dawn of a new Era

Applied Machine Learning 1:16:22

Applied Machine Learning 2019 - Lecture 09 - Gradient boosting; Calibration

Applied Machine Learning 1:15:04

Applied Machine Learning 2019 - Lecture 22 - Advanced Neural Networks

Applied Machine Learning 1:03:48

Applied Machine Learning 2019 - Lecture 11 - Imbalanced data

Applied Machine Learning 1:17:08

Applied Machine Learning 2019 - Lecture 05 - Preprocessing

Applied Machine Learning 1:17:39

Applied Machine Learning 2019 - Lecture 07 - Linear Models for Classifications, SVMs

Applied Machine Learning 1:17:12

Applied Machine Learning 2019 - Lecture 03 - Visualization and Matplotlib

Applied Machine Learning 1:17:05

Applied Machine Learning 2019 - Lecture 12 - Model Interpretration and Feature Selection

Applied Machine Learning 0:30:38

Applied Machine Learning - Peter Heiberg

Applied Machine Learning 1:18:29

Applied Machine Learning 2019 - Lecture 08 - Trees, Forests and Ensembles

Applied Machine Learning 1:17:55

Applied Machine Learning 2019 - Lecture 04 - Introduction to supervised learning

Applied Machine Learning 1:15:24

Applied Machine Learning 2019 - Lecture 14 - Dimensionality Reduction

Applied Machine Learning 1:16:57

Applied Machine Learning 2019 - Lecture 21 - Keras and Convolutional Neural Networks

Applied Machine Learning 0:40:32

Applied Machine Learning for Ranking Products in an Ecommerce Setting Arnoud de Munnik Wehkamp Jerry

Комментарии
Автор

And thank you soooo much for posting these video lectures online. I find them so helpful!

chrishassan
Автор

Really great lecture, I learnt a lot from it!

neptunesbounty
Автор

I Just had one question related to smote, how does it handle categorical data that is being generated. i.e. if one observation had a categorical variable that had a value hot encoded as 0 was being synthetically added to another observation that had it's value set as 1 and a value of 0.3 was used to connect the two points then the variable for the new observation would be 0.3. Because it's a categorical variable I don't get how that value is possible. Would SMOTE set that value to 0? as it's less than 0.5 or would it set it to 1? as it's greater than 0 or would it keep it as 0.3 when model building which would give an interesting interpretation of the value being synthesised.

chrishassan