Merge pull request #118 from ModelOriented/cran-candidate

CRAN release

.Rbuildignore

@@ -10,3 +10,4 @@
 ^test.R$
 ^backlog$
 ^CRAN-SUBMISSION$
+^pkgdown$

CRAN-SUBMISSION

@@ -1,3 +1,3 @@
-Version: 1.1.2
-Date: 2024-02-03 15:31:07 UTC
-SHA: 9eebdab2e583bf97d51dff9e7783df2b87751097
+Version: 1.2.0
+Date: 2024-07-12 12:10:00 UTC
+SHA: daf4cee64500abb8d78f92d8b1e8f1e588a59884

DESCRIPTION

@@ -30,5 +30,5 @@ Imports:
 Suggests: 
     testthat (>= 3.0.0)
 Config/testthat/edition: 3
-URL: https://github.com/ModelOriented/hstats, https://mayer79.github.io/hstats
-BugReports: https://github.com/ModelOriented/hstats/issues
+URL: https://github.com/ModelOriented/hstats/, https://modeloriented.github.io/hstats/
+BugReports: https://github.com/ModelOriented/hstats/issues/

NEWS.md

@@ -1,18 +1,14 @@
 # hstats 1.2.0
 
-## New home
+## My new home
 
 - My brand new home: https://github.com/ModelOriented/hstats
 
-## Major changes
+## Other changes
 
 - Factor-valued predictions are no longer possible.
 - Consequently, also removed "classification_error" loss.
 
-## Minor changes
-
-- Code simplifications.
-
 # hstats 1.1.2
 
 ## ICE plots
@@ -28,29 +24,29 @@
 
 ## Performance improvements
 
-- For pure data.frames (no tibbles, data.tables etc.), most functions are significantly faster ([#110](https://github.com/mayer79/hstats/pull/110)).
-- Slight speed-up of permutation importance for non-matrix `X` ([#109](https://github.com/mayer79/hstats/pull/109)).
+- For pure data.frames (no tibbles, data.tables etc.), most functions are significantly faster ([#110](https://github.com/ModelOriented/hstats/pull/110)).
+- Slight speed-up of permutation importance for non-matrix `X` ([#109](https://github.com/ModelOriented/hstats/pull/109)).
 
 ## Other changes
 
-- In multivariate cases, it was possible that normalized H-statistics could equal `0/0 (= NaN)`. Such values are now replaced by 0 ([#107](https://github.com/mayer79/hstats/issues/107)).
-- Removed an unnecessary special case when calculating column means ([#106](https://github.com/mayer79/hstats/pull/106)).
+- In multivariate cases, it was possible that normalized H-statistics could equal `0/0 (= NaN)`. Such values are now replaced by 0 ([#107](https://github.com/ModelOriented/hstats/issues/107)).
+- Removed an unnecessary special case when calculating column means ([#106](https://github.com/ModelOriented/hstats/pull/106)).
 
 # hstats 1.1.0
 
 ## Enhancements
 
-- {hstats} now also works for factor predictions. The levels are represented by one-hot-encoded columns ([PR#101](https://github.com/mayer79/hstats/pull/101)).
-- The plot method of a two-dimensional PDP has recieved the option `d2_geom = "line"`. Instead of a heatmap of the two features, one of the features is moved to color grouping. Combined with `swap_dim = TRUE`, you can swap the role of the two `v` variables without recalculating anything. The idea was proposed by [Roel Verbelen](https://github.com/RoelVerbelen) in [issue #91](https://github.com/mayer79/hstats/issues/91), see also [issue #94](https://github.com/mayer79/hstats/issues/94).
+- {hstats} now also works for factor predictions. The levels are represented by one-hot-encoded columns ([PR#101](https://github.com/ModelOriented/hstats/pull/101)).
+- The plot method of a two-dimensional PDP has recieved the option `d2_geom = "line"`. Instead of a heatmap of the two features, one of the features is moved to color grouping. Combined with `swap_dim = TRUE`, you can swap the role of the two `v` variables without recalculating anything. The idea was proposed by [Roel Verbelen](https://github.com/RoelVerbelen) in [issue #91](https://github.com/ModelOriented/hstats/issues/91), see also [issue #94](https://github.com/ModelOriented/hstats/issues/94).
 
 ## Bug fixes
 
-- Using `BY` and `w` via column names would fail for tibbles. This problem was described in [#92](https://github.com/mayer79/hstats/issues/92) by [Roel Verbelen](https://github.com/RoelVerbelen). Thx!
+- Using `BY` and `w` via column names would fail for tibbles. This problem was described in [#92](https://github.com/ModelOriented/hstats/issues/92) by [Roel Verbelen](https://github.com/RoelVerbelen). Thx!
 
 ## Other changes
 
-- Much faster one-hot-encoding, thanks to Mathias Ambühl ([PR#101](https://github.com/mayer79/hstats/pull/101)).
-- Most functions are slightly faster ([PR#101](https://github.com/mayer79/hstats/pull/101)).
+- Much faster one-hot-encoding, thanks to Mathias Ambühl ([PR#101](https://github.com/ModelOriented/hstats/pull/101)).
+- Most functions are slightly faster ([PR#101](https://github.com/ModelOriented/hstats/pull/101)).
 - Add unit tests to compare against {iml}.
 - Made all examples "tibble" and "data.table" friendly.
 - Revised input checks in loss functions (relevant for `perm_importance()` and `average_loss()`).

README.md

@@ -152,7 +152,8 @@ Let's study different plots to understand *how* the strong interaction between d
 They all reveal a substantial interaction between the two variables in the sense that the age effect gets weaker the closer to the ocean. Note that numeric `BY` features are automatically binned into quartile groups.
 
 ```r
-plot(partial_dep(fit, v = "age", X = X_train, BY = "log_ocean"), show_points = FALSE)
+partial_dep(fit, v = "age", X = X_train, BY = "log_ocean") |> 
+  plot(show_points = FALSE)
 ```
 
 ![](man/figures/pdp_ocean_age.svg)
@@ -167,8 +168,8 @@ plot(pd, d2_geom = "line", show_points = FALSE)
 ![](man/figures/pdp_2d_line.svg)
 
 ```r
-ic <- ice(fit, v = "age", X = X_train, BY = "log_ocean")
-plot(ic, center = TRUE)
+ice(fit, v = "age", X = X_train, BY = "log_ocean") |> 
+  plot(center = TRUE)
 ```
 
 ![](man/figures/ice.svg)
@@ -178,11 +179,14 @@ plot(ic, center = TRUE)
 In the spirit of [1], and related to [4], we can extract from the "hstats" objects a partial dependence based variable importance measure. It measures not only the main effect strength (see [4]), but also all its interaction effects. It is rather experimental, so use it with care (details in the section "Background"):
 
 ```r
-plot(pd_importance(s))
+pd_importance(s) |> 
+  plot()
 
 # Compared with four times repeated permutation importance regarding MSE
 set.seed(10)
-plot(perm_importance(fit, X = X_valid, y = y_valid))
+
+perm_importance(fit, X = X_valid, y = y_valid) |> 
+  plot()
 ```
 
 ![](man/figures/importance.svg)
@@ -209,14 +213,17 @@ s <- hstats(ex)
 s  # 0.054
 plot(s)
 
-# Strongest relative interaction
-plot(ice(ex, v = "Sepal.Width", BY = "Petal.Width"), center = TRUE)
-plot(partial_dep(ex, v = "Sepal.Width", BY = "Petal.Width"), show_points = FALSE)
-plot(partial_dep(ex, v = c("Sepal.Width", "Petal.Width"), grid_size = 200))
+# Strongest relative interaction (different visualizations)
+ice(ex, v = "Sepal.Width", BY = "Petal.Width") |> 
+  plot(center = TRUE)
+
+partial_dep(ex, v = "Sepal.Width", BY = "Petal.Width") |> 
+  plot(show_points = FALSE)
 
-perm_importance(ex)
+partial_dep(ex, v = c("Sepal.Width", "Petal.Width"), grid_size = 200) |> 
+  plot()
 
-# Permutation importance
+perm_importance(ex)
 # Petal.Length  Petal.Width  Sepal.Width      Species 
 #   0.59836442   0.11625137   0.07966910   0.03982554 
 ```
@@ -298,7 +305,7 @@ fit <- lgb.train(
 average_loss(fit, X = X_valid, y = y_valid, loss = "mlogloss")
 
 perm_importance(fit, X = X_valid, y = y_valid, loss = "mlogloss", m_rep = 100)
-# Permutation importance regarding mlogloss
+
 # Petal.Length  Petal.Width  Sepal.Width Sepal.Length 
 #  2.624241332  1.011168660  0.082477177  0.009757393
 
@@ -396,7 +403,9 @@ fit <- iris_wf |>
 
 s <- hstats(fit, X = iris[, -1])
 s # 0 -> no interactions
-plot(partial_dep(fit, v = "Petal.Width", X = iris))
+
+partial_dep(fit, v = "Petal.Width", X = iris) |> 
+  plot()
 
 imp <- perm_importance(fit, X = iris, y = "Sepal.Length")
 imp
@@ -425,8 +434,11 @@ fit <- train(
 
 h2(hstats(fit, X = iris[, -1]))  # 0
 
-plot(ice(fit, v = "Petal.Width", X = iris), center = TRUE)
-plot(perm_importance(fit, X = iris, y = "Sepal.Length"))
+ice(fit, v = "Petal.Width", X = iris) |> 
+  plot(center = TRUE)
+
+perm_importance(fit, X = iris, y = "Sepal.Length") |> 
+  plot()
 ```
 
 ### mlr3

cran-comments.md

@@ -1,23 +1,19 @@
-# hstats 1.1.2
+# Re-submission: hstats 1.2.0
 
-Hello CRAN team
+Moving the github repo has left some old links in the NEWS file. This is fixed here.
 
-This is a small release with two convenient API improvements.
+# Original message
 
-## Local checks: 0 errors, 0 warnings, 0 notes
+Hello CRAN
 
-## Rhub: 3 NOTES (sounding harmless)
+This release mainly updates the new repository ("ModelOriented" of TU Warcaw instead of my personal one), and adds Prof Biecek as co-author.
 
-- checking HTML version of manual ... NOTE
-  Skipping checking math rendering: package 'V8' unavailable
-- checking for non-standard things in the check directory ... NOTE
-  Found the following files/directories:
-  ''NULL''
-- checking for detritus in the temp directory ... NOTE
-  Found the following files/directories:
-  'lastMiKTeXException'
+## Local checks
+
+0 errors, 0 warnings, 0 notes
 
 ## Winbuilder
 
-Status: OK
+Status: 1 NOTE
+R Under development (unstable) (2024-07-11 r86890 ucrt)