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

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") ```