Understanding model interpretability in R with ggplot2 and mikropml (CC134)

The interpretability of a machine learning model tends to vary by the performance of the model. The need to interpret your model depends on what you hope to do with that model. In this Code Club, Pat shows how you can extract interpretability data from models created using mikropml and visualize the importance of features that are used in the model.

Pat will use functions from the #mikropml R package and the #ggplot2 and #dplyr packages in #RStudio. The accompanying blog post can be found at https://www.riffomonas.org/code_club/....


If you're interested in taking an upcoming 3 day R workshop, email me at [email protected]!

R: https://r-project.org
RStudio: https://rstudio.com
Raw data: https://github.com/riffomonas/raw_dat...
Workshops: https://www.mothur.org/wiki/workshops

You can also find complete tutorials for learning R with the tidyverse using...
Microbial ecology data: https://www.riffomonas.org/minimalR/
General data: https://www.riffomonas.org/generalR/

0:00 Introduction
4:34 Extracting weights from L2 regularlized logistic regression
9:24 How to convert a matrix into a tibble
12:08 How to read in 100 files with map_dfr
13:41 Using stat_summary to plot the weights
15:41 Creating point range plot without stat_summary
17:43 Sorting and filtering data in figure
19:55 Getting data from permutation importance test
24:51 Modifying weights plot for permutation importance
26:28 Recap