The modelStudio package automates explanation of machine learning predictive models. This package generates advanced interactive and animated model explanations in the form of a serverless HTML site.

It combines R with D3.js to produce plots and descriptions for various local and global explanations. Tools for model exploration unite with tools for EDA (Exploratory Data Analysis) to give a broad overview of the model behaviour. modelStudio is a fast and condensed way to get all the answers without much effort. Break down your model and look into its ingredients with only a few lines of code.

See a demo   explainFIFA19   More

The modelStudio package is a part of the DrWhy.AI universe.

# Install from CRAN: 
install.packages("modelStudio")

# Install the development version from GitHub:
devtools::install_github("ModelOriented/modelStudio")

This package bases on DALEX explainers created with DALEX::explain().

library("modelStudio")

# Create a model
model <- glm(survived ~.,
             data = DALEX::titanic_imputed,
             family = "binomial")
                 
# Wrap it into an explainer        
explainer <- DALEX::explain(model,
                            data = DALEX::titanic_imputed[,-8],
                            y = DALEX::titanic_imputed[,8],
                            label = "glm")
                   
# Pick some data points
new_observations <- DALEX::titanic_imputed[1:4,]
rownames(new_observations) <- c("Lucas", "James", "Thomas", "Nancy")

# Make a studio for the model
modelStudio(explainer, new_observations)

More examples here.

• Conference Poster about modelStudio

• Article about modelStudio

• News

• Read the vignette: modelStudio - perks and features

• Cheatsheet

Save modelStudio as a HTML file using buttons on the top of the RStudio Viewer or with the r2d3::save_d3_html() function.

Work on this package was financially supported by the NCN Opus grant 2016/21/B/ST6/02176.