Learn R Programming

DALEX (version 2.4.3)

moDel Agnostic Language for Exploration and eXplanation

Description

Any unverified black box model is the path to failure. Opaqueness leads to distrust. Distrust leads to ignoration. Ignoration leads to rejection. DALEX package xrays any model and helps to explore and explain its behaviour. Machine Learning (ML) models are widely used and have various applications in classification or regression. Models created with boosting, bagging, stacking or similar techniques are often used due to their high performance. But such black-box models usually lack direct interpretability. DALEX package contains various methods that help to understand the link between input variables and model output. Implemented methods help to explore the model on the level of a single instance as well as a level of the whole dataset. All model explainers are model agnostic and can be compared across different models. DALEX package is the cornerstone for 'DrWhy.AI' universe of packages for visual model exploration. Find more details in (Biecek 2018) .

Copy Link

Version

Install

install.packages('DALEX')

Monthly Downloads

6,097

Version

2.4.3

License

GPL

Homepage

https://modeloriented.github.io/DALEX/

Maintainer

Przemyslaw Biecek

Last Published

January 15th, 2023

Functions in DALEX (2.4.3)

Wrapper for Loss Functions from the yardstick Package

Exract info from model

Plot List of Explanations

model_performance

Dataset Level Model Performance Measures

plot.model_diagnostics

Plot Dataset Level Model Diagnostics

Dataset Level Variable Profile as Partial Dependence or Accumulated Local Dependence Explanations

model_diagnostics

Dataset Level Model Diagnostics

Dataset Level Variable Importance as Change in Loss Function after Variable Permutations

plot.model_parts

Plot Variable Importance Explanations

plot.model_performance

Plot Dataset Level Model Performance Explanations

predict_diagnostics

Instance Level Residual Diagnostics

print.description

Print Natural Language Descriptions

predict_profile

Instance Level Profile as Ceteris Paribus

plot.predict_profile

Plot Variable Profile Explanations

plot.predict_parts

Plot Variable Attribution Explanations

Instance Level Parts of the Model Predictions

plot.predict_diagnostics

Plot Instance Level Residual Diagnostics

set_theme_dalex

Default Theme for DALEX plots

print.explainer

Print Explainer Summary

plot.model_profile

Plot Dataset Level Model Profile Explanations

print.model_diagnostics

Print Dataset Level Model Diagnostics

shap_aggregated

SHAP aggregated values

Wrap Various Predict Functions

plot.shap_aggregated

Plot Generic for Break Down Objects

variable_effect

Dataset Level Variable Effect as Partial Dependency Profile or Accumulated Local Effects

predict.explainer

Predictions for the Explainer

print.model_profile

Print Dataset Level Model Profile

Passengers and Crew on the RMS Titanic Data

print.predict_diagnostics

Print Instance Level Residual Diagnostics

DrWhy Theme for ggplot objects

Update data of an explainer object

print.model_performance

Print Dataset Level Model Performance Summary

print.model_info

Print model_info

Update label of explainer object

loss_cross_entropy

Calculate Loss Functions

install_dependencies

Install all dependencies for the DALEX package

Data for early COVID mortality

colors_discrete_drwhy

DrWhy color palettes for ggplot objects

explain.default

Create Model Explainer

Human Resources Data

Apartments Data

World Happiness Report data

FIFA 20 preprocessed data