Merge pull request #56 from mayer79/v2_submission

CRAN version 0.2.0

CRAN-SUBMISSION

@@ -1,3 +1,3 @@
-Version: 0.1.0
-Date: 2023-07-16 18:48:26 UTC
-SHA: 9638974266fcc5f32a51870bccf0d08bde810551
+Version: 0.2.0
+Date: 2023-09-03 16:23:41 UTC
+SHA: 9fb4db1361c4593b956fb65fd122ed49d1b01caa

NEWS.md

@@ -15,7 +15,7 @@ Note that such a custom function needs to return per-row losses, not their avera
 ## Minor changes
 
 - `summary.hstats()` now returns an object of class "summary_hstats" with its own `print()` method. Like this, one can use `su <- summary()` without printing to the console.
-- The output of `summary.hstats()` is printed sligthly more compact.
+- The output of `summary.hstats()` is printed slightly more compact.
 - `plot.hstats()` has recieved a `rotate_x = FALSE` argument for rotating x labels by 45 degrees.
 - `plot.hstats()` and `summary.hstats()` have received explicit arguments `normalize`, `squared`, `sort`, `eps` instead of passing them via `...`.
 - `plot.hstats()` now passes `...` to `geom_bar()`.

R/hstats.R

@@ -35,12 +35,12 @@
 #' @param w Optional vector of case weights for each row of `X`.
 #' @param pairwise_m Number of features for which pairwise statistics are to be 
 #'   calculated. The features are selected based on Friedman and Popescu's overall 
-#'   interaction strength \eqn{H^2_j}. 
-#'   Set to `length(v)` to calculate every pair and to 0 to avoid pairwise calculations.
+#'   interaction strength \eqn{H^2_j}. Set to to 0 to avoid pairwise calculations.
 #'   For multivariate predictions, the union of the column-wise strongest variable
 #'   names is taken. This can lead to very long run-times.
 #' @param threeway_m Same as `pairwise_m`, but controlling the number of features for
 #'   which threeway interactions should be calculated. Not larger than `pairwise_m`.
+#'   Set to 0 to avoid threeway calculations.
 #' @param verbose Should a progress bar be shown? The default is `TRUE`.
 #' @param ... Additional arguments passed to `pred_fun(object, X, ...)`, 
 #'   for instance `type = "response"` in a [glm()] model.

cran-comments.md

@@ -1,34 +1,20 @@
-# Resubmission
-
-- Fixing an indirect URL in the README
-- Sticking to "authors (year) <doi>" reference in DESCRIPTION.
-
-# Original message
+# hstats 0.2.0
 
 Hello CRAN team
 
-Trying to submit a new package that calculates Friedman and Popescu's H statistics in many variants.
-
-If we can't fit it before your summer break - no hurry :-).
+{hstats} received a fast, multivariate permutation importance function, and some API improvements. Not yet ready for version 1.0.0, but almost :-).
 
-Thanks for all your hard work
+Thanks a lot for caring about R so much
 
 Michael
 
-## Local checks seem ok
-
-❯ checking HTML version of manual ... NOTE
-  Skipping checking HTML validation: no command 'tidy' found
+## Local checks: OK
 
-## Rhub fires some false alarms on spelling mistakes
-
-New submission
-
-Possibly misspelled words in DESCRIPTION:
-  explainers (13:32)
+## Rhub has its typical notes
 
-## Winbuilder seems ok
+* checking HTML version of manual ... NOTE
+Skipping checking HTML validation: no command 'tidy' found
+Skipping checking math rendering: package 'V8' unavailable
 
-Status: 1 NOTE
-R Under development (unstable) (2023-07-15 r84693 ucrt)
+## Winbuilder: OK