pdp <img src="man/figures/pdp-logo.png" align="right" width="130" height="150" />
=================================================================================
[](https://cran.r-project.org/package=pdp)
[](https://github.com/bgreenwell/pdp/actions)
[](https://codecov.io/gh/bgreenwell/pdp?branch=master)
[](http://cranlogs.r-pkg.org/badges/grand-total/pdp)
[](https://www.tidyverse.org/lifecycle/#stable)
Overview
--------
[pdp](https://cran.r-project.org/package=pdp) is an R package for
constructing ***p**artial **d**ependence **p**lots* (PDPs) and
***i**ndividual **c**onditional **e**xpectation* (ICE) curves. PDPs and
ICE curves are part of a larger framework referred to as *interpretable
machine learning* (IML), which also includes (but not limited to)
***v**ariable **i**mportance **p**lots* (VIPs). While VIPs (available in
the R package [vip](https://koalaverse.github.io/vip/index.html)) help
visualize feature impact (either locally or globally), PDPs and ICE
curves help visualize feature effects. An in-progress, but
comprehensive, overview of IML can be found at the following URL:
<a href="https://github.com/christophM/interpretable-ml-book" class="uri">https://github.com/christophM/interpretable-ml-book</a>.
A detailed introduction to [pdp](https://cran.r-project.org/package=pdp)
has been published in The R Journal: “pdp: An R Package for Constructing
Partial Dependence Plots”,
<a href="https://journal.r-project.org/archive/2017/RJ-2017-016/index.html" class="uri">https://journal.r-project.org/archive/2017/RJ-2017-016/index.html</a>.
You can track development at
<a href="https://github.com/bgreenwell/pdp" class="uri">https://github.com/bgreenwell/pdp</a>.
To report bugs or issues, contact the main author directly or submit
them to
<a href="https://github.com/bgreenwell/pdp/issues" class="uri">https://github.com/bgreenwell/pdp/issues</a>.
For additional documentation and examples, visit the [package
website](https://bgreenwell.github.io/pdp/index.html).
As of right now, `pdp` exports the following functions:
- `partial()` - compute partial dependence functions and individual
conditional expectations (i.e., objects of class `"partial"` and
`"ice"`, respectively) from various fitted model objects;
- `plotPartial()"` - construct `lattice`-based PDPs and ICE curves;
- `autoplot()` - construct `ggplot2`-based PDPs and ICE curves;
- ~~`topPredictors()` extract most “important” predictors from various
types of fitted models.~~ see
[vip](https://koalaverse.github.io/vip/index.html) instead for a
more robust and flexible replacement;
- `exemplar()` - construct an exemplar record from a data frame
(**experimental** feature that may be useful for constructing fast,
approximate feature effect plots.)
Installation
------------
``` r
# The easiest way to get pdp is to install it from CRAN:
install.packages("pdp")
# Alternatively, you can install the development version from GitHub:
if (!("remotes" %in% installed.packages()[, "Package"])) {
install.packages("remotes")
}
remotes::install_github("bgreenwell/pdp")
```
pdp：从R中的各种类型的机器学习模型构造部分依赖（即边际效应）图的通用框架
5星 · 超过95%的资源 需积分: 49 65 浏览量
2021-02-05
14:46:53
上传
评论 4
收藏 13.02MB ZIP 举报 
pdp：从R中的各种类型的机器学习模型构造部分依赖（即边际效应）图的通用框架 （241个子文件） 
298C9769 32B 2A4095D0 3B 3FBFDEDC 32B 4E70B8D6 3B 7974D559 32B 8B513ADF 3B __packages 38B B31D7B02 3B BE222925 3B greenwell.bib 12KB pdp-pkg.bib 10KB build_options 199B in_out.c 3KB pdp-init.c 627B C1EAEEEF 32B CITATION 693B comments 8KB PartialGBM.cpp 6KB cpp-definition-cache 2KB docsearch.css 11KB pkgdown.css 4KB california_housing.csv 1.64MB DESCRIPTION 1KB pdp.dll 18KB .DS_Store 14KB .DS_Store 6KB .DS_Store 6KB .DS_Store 6KB .gitignore 101B .gitignore 7B
pdp-intro.synctex.gz 1KB PartialGBM.h 679B pdp.h 163B
response.html 810KB response.nb.html 810KB response_reviewer_1.nb.html 788KB response_reviewer_2.nb.html 770KB pdp.html 65KB pdp-example-tensorflow.html 38KB pdp-computation.html 34KB pdp-extending.html 34KB computation.html 32KB partial.html 31KB pdp-classification.html 22KB pdp-example-xgboost.html 20KB index.html 19KB autoplot.partial.html 19KB plotPartial.html 18KB pdp-se.Rmd.html 18KB pdp-reticulate.html 17KB index.html 12KB boston.html 10KB pima.html 9KB topPredictors.html 9KB pdp.html 8KB index.html 7KB progress_progress.html 6KB authors.html 6KB grid.arrange.html 6KB trellis.last.object.html 6KB index.html 6KB pipe.html 6KB extending.html 5KB parallel.html 5KB INDEX 12KB pkgdown.js 3KB docsearch.js 2KB pdp-intro.log 22KB NEWS.md 10KB README.md 3KB TODO.md 766B xgboost.model 99KB NAMESPACE 2KB
pdp-intro.pdf 823KB pdp-intro.pdf 815KB response.pdf 229KB response_reviewer_1.pdf 186KB response_reviewer_2.pdf 177KB persistent-state 2KB
README-example-svm-2-1.png 543KB README-example-svm-2-1.png 543KB ice-curves-1.png 410KB ice-curves-1.png 410KB ice-1.png 367KB README-example-rf-3-1.png 322KB README-example-rf-3-1.png 322KB iris-svm-pdp-logit-1.png 317KB boston-ranger-ice-curves-1.png 295KB boston-ranger-cice-curves-1.png 287KB all-results-1.png 237KB all-results-1.png 237KB unnamed-chunk-2-1.png 166KB ames-xgb-pdp-1.png 153KB boston-rf-03-1.png 153KB boston-rf-03-1.png 153KB unnamed-chunk-5-1.png 143KB README-example-rf-1.png 111KB README-example-rf-1.png 111KB README-example-rf-2-1.png 105KB README-example-rf-2-1.png 105KB共 241 条
- 1
- 2
- 3
评论1