XAI Stories. Case studies for eXplainable Artificial Intelligence

ebook: https://pbiecek.github.io/xai_stories/

Machine learning has a number of applications. Very often, however, machine learning predictive models are treated as black boxes which can be automatically trained without worrying about the domain in which they are used.

The development of XAI tools allows to x-ray these black boxes. In the following chapters, we show example applications of different XAI techniques to real-world problems.

How this book came about

This book is the result of a student projects for Interpretable Machine Learning course at the University of Warsaw and the Warsaw University of Technology. Each team has prepared one case study for selected XAI technique.

This project is inspired by a fantastic book Limitations of Interpretable Machine Learning Methods done at the Department of Statistics, LMU Munich. We used the LIML project as the cornerstone for this repository.

How to build the book

Step 1: Clone or download the repository https://github.com/pbiecek/xai_stories.

Step 2: Install dependencies

devtools::install_dev_deps()

Step 3: Render the book (R commands)

bookdown::render_book('./', 'bookdown::gitbook')

The book does not build

OS: Ubuntu 20.04 LTS
R version: 3.6.3 (2020-02-29) -- "Holding the Windsock"
pandoc version:

$ pandoc --version                                                                                              (git)-[master] 
pandoc 2.9.2
Compiled with pandoc-types 1.20, texmath 0.12.0.1, skylighting 0.8.3.2
Default user data directory: /home/witalis/.local/share/pandoc or /home/witalis/.pandoc
Copyright (C) 2006-2019 John MacFarlane
Web:  https://pandoc.org
This is free software; see the source for copying conditions.
There is no warranty, not even for merchantability or fitness
for a particular purpose.

sessionInfo (after the devtools::install_dev_deps() command):

> sessionInfo()
R version 3.6.3 (2020-02-29)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 20.04 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.9.0
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.9.0

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=pl_PL.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=pl_PL.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=pl_PL.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=pl_PL.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

loaded via a namespace (and not attached):
 [1] ps_1.3.3          fansi_0.4.1       prettyunits_1.1.1 rprojroot_1.3-2  
 [5] withr_2.2.0       digest_0.6.25     crayon_1.3.4      assertthat_0.2.1 
 [9] R6_2.4.1          backports_1.1.7   magrittr_1.5      rlang_0.4.6      
[13] cli_2.0.2         fs_1.4.1          remotes_2.1.1     testthat_2.3.2   
[17] callr_3.4.3       ellipsis_0.3.1    desc_1.2.0        devtools_2.3.0   
[21] tools_3.6.3       glue_1.4.1        pkgload_1.1.0     compiler_3.6.3   
[25] processx_3.4.2    pkgbuild_1.0.8    sessioninfo_1.1.1 memoise_1.1.0    
[29] usethis_1.6.1

Steps to reproduce

Step 1: Clone or download the repository https://github.com/pbiecek/xai_stories.

Step 2: Install dependencies

devtools::install_dev_deps()

Step 3: Render the book (R commands)

bookdown::render_book('./', 'bookdown::gitbook')

Result

Quitting from lines 2197-2199 (book.Rmd) 
Error in knitr::include_graphics("images/09_recency.PNG") : 
  Cannot find the file(s): "images/09_recency.PNG"

Possible steps to fix

After creating a images/09_recency.PNG file, with the same steps:

Quitting from lines 2197-2199 (book.Rmd) 
Error in knitr::include_graphics("images/09_history.PNG") : 
  Cannot find the file(s): "images/09_history.PNG"

After creating a images/09_history.PNG file, with the same steps:

Quitting from lines 2220-2222 (book.Rmd) 
Error in knitr::include_graphics("images/09_uplift_train.png") : 
  Cannot find the file(s): "images/09_uplift_train.png"

After creating a images/09_uplift_train.png file, with the same steps:

Quitting from lines 2220-2222 (book.Rmd) 
Error in knitr::include_graphics("images/09_uplift_test.png") : 
  Cannot find the file(s): "images/09_uplift_test.png"

After creating a images/09_uplift_test.png file, with the same steps:

Quitting from lines 2280-2282 (book.Rmd) 
Error in knitr::include_graphics("images/09_shap_min_uplift.png") : 
  Cannot find the file(s): "images/09_shap_min_uplift.png"

After creating a images/09_shap_min_uplift.png file, with the same steps:

Quitting from lines 2280-2282 (book.Rmd) 
Error in knitr::include_graphics("images/09_shap_max_uplift.png") : 
  Cannot find the file(s): "images/09_shap_max_uplift.png"

After creating a images/09_shap_max_uplift.png file, with the same steps the build is succesfill (with some warnings)

pandoc-citeproc: reference 4groups_chart not found
pandoc-citeproc: reference pyuplift not found
pandoc-citeproc: reference pdp not found
pandoc-citeproc: reference shap_dependence_plot not found
pandoc-citeproc: reference shap_dependence_plot_explained not found

Additional context

Additionally files:

images/09_history.png
images/09_recency.png
images/09_shap_max_uplift.PNG
images/09_shap_min_uplift.PNG
images/09_uplift_test.PNG
images/09_uplift_train.PNG

are not used in the build.

Possible fix

Based on this information I had, I submitted a PR, namely:

        renamed:    images/09_history.png -> images/09_history.PNG
        renamed:    images/09_recency.png -> images/09_recency.PNG
        renamed:    images/09_shap_max_uplift.PNG -> images/09_shap_max_uplift.png
        renamed:    images/09_shap_min_uplift.PNG -> images/09_shap_min_uplift.png
        renamed:    images/09_uplift_test.PNG -> images/09_uplift_test.png
        renamed:    images/09_uplift_train.PNG -> images/09_uplift_train.png

Unfortunately the PR was criticized by the creators of the 09_*.{PNG,png} files.

I would be grateful for fixing this problem.

pbiecek / xai_stories Goto Github PK

xai_stories's Introduction

XAI Stories. Case studies for eXplainable Artificial Intelligence

How this book came about

How to build the book

xai_stories's People

Contributors

Stargazers

Watchers

Forkers

xai_stories's Issues

Steps to reproduce

Step 1: Clone or download the repository https://github.com/pbiecek/xai_stories.

Step 2: Install dependencies

Step 3: Render the book (R commands)

Result

Possible steps to fix

Additional context

Possible fix

Recommend Projects

Recommend Topics

Recommend Org