From e5924771d10ccb311070163e23a96e919ed46434 Mon Sep 17 00:00:00 2001
From: "Afkhami, Navid" <Navid.Afkhami@mdc-berlin.de>
Date: Fri, 19 Jul 2024 10:51:42 +0000
Subject: gnu: Add r-shapviz.

* gnu/packages/cran.scm (r-shapviz): New variable.

Change-Id: I9795dceda2534663b640a0fa936b10a4a2ff8fae
Signed-off-by: Ricardo Wurmus <rekado@elephly.net>
---
 gnu/packages/cran.scm | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

(limited to 'gnu')

diff --git a/gnu/packages/cran.scm b/gnu/packages/cran.scm
index 66205781b1..43e4cf416b 100644
--- a/gnu/packages/cran.scm
+++ b/gnu/packages/cran.scm
@@ -31863,6 +31863,42 @@ Mardia, K.V. (2016).  Statistical shape analysis, with Applications in R (2nd
 Edition), John Wiley and Sons.")
     (license license:gpl2)))
 
+(define-public r-shapviz
+  (package
+    (name "r-shapviz")
+    (version "0.9.3")
+    (source
+     (origin
+       (method url-fetch)
+       (uri (cran-uri "shapviz" version))
+       (sha256
+        (base32 "01zq6kr5na40xgypif6sv2pjynhva9v0npx779k7fxck1db68z80"))))
+    (properties `((upstream-name . "shapviz")))
+    (build-system r-build-system)
+    (propagated-inputs (list r-ggfittext
+                             r-gggenes
+                             r-ggplot2
+                             r-ggrepel
+                             r-patchwork
+                             r-rlang
+                             r-xgboost))
+    (native-inputs (list r-knitr))
+    (home-page "https://github.com/ModelOriented/shapviz")
+    (synopsis "SHAP visualizations")
+    (description
+     "This package provides visualizations for SHAP (SHapley Additive
+exPlanations) such as waterfall plots, force plots, various types of
+importance plots, dependence plots, and interaction plots.  These plots act on
+a @code{shapviz} object created from a matrix of SHAP values and a
+corresponding feature dataset.  Wrappers for the R packages @code{xgboost},
+@code{lightgbm}, @code{fastshap}, @code{shapr}, @code{h2o}, @code{treeshap},
+@code{DALEX}, and @code{kernelshap} are added for convenience.  By separating
+visualization and computation, it is possible to display factor variables in
+graphs, even if the SHAP values are calculated by a model that requires
+numerical features.  The plots are inspired by those provided by the
+@code{shap} package in Python, but there is no dependency on it.")
+    (license license:gpl2+)))
+
 (define-public r-animation
   (package
     (name "r-animation")
-- 
cgit v1.2.3