Release v0.10.0 (#224)

* Add slash to name * Fix Pareto/NBD print name (#201) * Update author name * Formula interface (#203) * Github actions: Upgrade to v2 (#214) * clvdata: Plot Transaction Timings (#213) * Housekeeping: Makevars (#217) * Run tests in parallel (#216) * Draw other models in same plot (#215) * PNBD Dyncov Rcpp (#205) * Update docu: continuous discount rate (#223) * Drop explicit c++11 specification * Prepare release * Fix CRAN CPU usage & fix docu
bachmannpatrick · Oct 23, 2023 · 203d218 · 203d218
2 parents 7b2cebb + bdfe92b
commit 203d218
Show file tree

Hide file tree

Showing 137 changed files with 6,041 additions and 2,704 deletions.
diff --git a/.github/workflows/Coverage.yaml b/.github/workflows/Coverage.yaml
@@ -8,13 +8,13 @@ jobs:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
 
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
 
-      - uses: r-lib/actions/setup-r@v1
+      - uses: r-lib/actions/setup-r@v2
         with:
           use-public-rspm: true
 
-      - uses: r-lib/actions/setup-r-dependencies@v1
+      - uses: r-lib/actions/setup-r-dependencies@v2
         with:
           extra-packages: covr
 

diff --git a/.github/workflows/R-CMD-check.yaml b/.github/workflows/R-CMD-check.yaml
@@ -27,45 +27,37 @@ jobs:
         R_KEEP_PKG_SOURCE: yes
 
     steps:
-        - uses: actions/checkout@v2
+        - uses: actions/checkout@v3
 
-        - uses: r-lib/actions/setup-pandoc@v1
+        - uses: r-lib/actions/setup-pandoc@v2
 
-        - uses: r-lib/actions/setup-r@v1
+        - uses: r-lib/actions/setup-r@v2
           with:
             r-version: ${{ matrix.config.r }}
             http-user-agent: ${{ matrix.config.http-user-agent }}
             use-public-rspm: true
 
-        - name: Install linux system dependencies
-          if: runner.os == 'Linux'
-          env:
-            RHUB_PLATFORM: linux-x86_64-ubuntu-gcc
-          run: |
-            sudo apt-get update
-            sudo apt-get install -y qpdf texlive-xetex ghostscript
-
         - name: Install mac system dependencies
           if: runner.os == 'macOS'
           run: |
-            brew install gsl openssl ghostscript
+            brew install gsl
 
         - name: Install Windows system dependencies
           if: runner.os == 'Windows'
           run: |
             pacman -Sy # Update repo index
             pacman -S --noconfirm mingw-w64-x86_64-gsl mingw-w64-i686-gsl mingw-w64-ucrt-x86_64-gsl
-            # pacman -S --noconfirm mingw-w64-{i686,x86_64}-gsl
 
-        - uses: r-lib/actions/setup-r-dependencies@v1
+        - uses: r-lib/actions/setup-r-dependencies@v2
           with:
-            extra-packages: rcmdcheck
+            extra-packages: any::rcmdcheck
+            needs: check
 
-        - uses: r-lib/actions/setup-tinytex@v1
+        - uses: r-lib/actions/setup-tinytex@v2
 
-        - uses: r-lib/actions/check-r-package@v1
+        - uses: r-lib/actions/check-r-package@v2
           with:
-            args: '"--no-manual", "--as-cran", "--no-tests"'
+            args: 'c("--no-manual", "--as-cran", "--no-tests")'
 
         - name: Upload check results
           if: failure()

diff --git a/.github/workflows/Tests.yaml b/.github/workflows/Tests.yaml
@@ -12,27 +12,39 @@ on:
 
 jobs:
   tests-ubuntu-latest:
-    runs-on: ubuntu-latest
+    runs-on: macOS-latest
     env:
       R_REMOTES_NO_ERRORS_FROM_WARNINGS: true
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
 
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
 
-      - uses: r-lib/actions/setup-r@v1
+      - uses: r-lib/actions/setup-r@v2
         with:
           use-public-rspm: true
 
-      - uses: r-lib/actions/setup-r-dependencies@v1
+      - name: Install mac system dependencies
+        run: |
+          brew install gsl
+
+      - uses: r-lib/actions/setup-r-dependencies@v2
         with:
           extra-packages: devtools
 
+      - name: diag output
+        run: sysctl hw.logicalcpu
+
       - name: Run tests
         run: |
+
           library(testthat)
           library(devtools)
 
+          # Ncpus=1 by default, and takes precedence over TESTTHAT_CPUS env var
+          # set to 3 which is default num cpus on mac GH actions runner
+          options(Ncpus=3)
+
           # If on a release branch (name starts with 'release'),
           #   run all tests, including lengthy ones marked with skip_on_ci()
           if(grepl(pattern = "^release", x = Sys.getenv("GITHUB_HEAD_REF"))){

diff --git a/.github/workflows/pkgdown.yaml b/.github/workflows/pkgdown.yaml
@@ -10,38 +10,22 @@ jobs:
     env:
       GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v3
 
-      - uses: r-lib/actions/setup-pandoc@master
+      - uses: r-lib/actions/setup-pandoc@v2
 
-      - uses: r-lib/actions/setup-r@master
-
-      - name: Install dependencies for mac
-        run: |
-          brew install gsl openssl
-
-      - name: Query dependencies
-        run: |
-          install.packages('remotes')
-          saveRDS(remotes::dev_package_deps(dependencies = TRUE), ".github/depends.Rds", version = 2)
-        shell: Rscript {0}
-
-      - name: Cache R packages
-        uses: actions/cache@v1
+      - uses: r-lib/actions/setup-r@v2
         with:
-          path: ${{ env.R_LIBS_USER }}
-          key: macOS-r-4.0-0-${{ hashFiles('.github/depends.Rds') }}
-          restore-keys: macOS-r-4.0-0-
+          use-public-rspm: true
 
-      - name: Install dependencies
+      - name: Install mac system dependencies
         run: |
-          install.packages("remotes")
-          remotes::install_deps(dependencies = TRUE)
-          remotes::install_dev("pkgdown")
-        shell: Rscript {0}
+          brew install gsl
 
-      - name: Install package
-        run: R CMD INSTALL .
+      - uses: r-lib/actions/setup-r-dependencies@v2
+        with:
+          extra-packages: any::pkgdown, local::.
+          needs: website
 
       - name: Deploy package
         run: pkgdown::deploy_to_branch(new_process = FALSE)

diff --git a/CLVTools.Rproj b/CLVTools.Rproj
@@ -17,6 +17,7 @@ StripTrailingWhitespace: Yes
 
 BuildType: Package
 PackageUseDevtools: Yes
+PackageCleanBeforeInstall: No
 PackageInstallArgs: --no-multiarch --with-keep.source
 PackageCheckArgs: --no-tests --as-cran --no-vignettes --no-build-vignettes --ignore-vignettes
 PackageRoxygenize: rd,collate,namespace
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,13 +1,13 @@
 Package: CLVTools
 Title: Tools for Customer Lifetime Value Estimation
-Version: 0.9.0
-Date: 2022-01-07
+Version: 0.10.0
+Date: 2023-10-23
 Authors@R: c(
     person(given="Patrick", family="Bachmann",    email = "[email protected]", role = c("cre","aut")),
     person(given="Niels",   family="Kuebler",     email = "[email protected]", role = "aut"),
     person(given="Markus",  family="Meierer",     email = "[email protected]", role = "aut"),
     person(given="Jeffrey", family="Naef",        email = "[email protected]",role = "aut"),
-    person(given="Elliot",  family="Oblander",    email = "[email protected]",role = "aut"),
+    person(given="E. Shin", family="Oblander",    email = "[email protected]",role = "aut"),
     person(given="Patrik",  family="Schilter",    email = "[email protected]",role = "aut")
     )
 Depends:
@@ -33,6 +33,7 @@ Description:
     Further, we provide an implementation of the Gamma/Gamma model to model the spending process of individuals.
 Imports:
     data.table (>= 1.12.0),
+    Formula (>= 1.2-4),
     ggplot2 (>= 3.2.0),
     lubridate (>= 1.7.8),
     Matrix (>= 1.2-17),
@@ -45,19 +46,20 @@ Suggests:
     covr,
     knitr,
     rmarkdown,
-    testthat
+    xml2,
+    testthat (>= 3.0.0)
 License: GPL-3
 URL: https://github.com/bachmannpatrick/CLVTools
 BugReports: https://github.com/bachmannpatrick/CLVTools/issues
 NeedsCompilation: yes
-SystemRequirements: C++11
-LinkingTo: Rcpp, RcppArmadillo (>= 0.9.500.2.0), RcppGSL (>= 0.3.7)
+LinkingTo: Rcpp, RcppArmadillo (>= 0.9.500.2.0), RcppGSL (>= 0.3.7), testthat
 LazyLoad: yes
 Encoding: UTF-8
 Collate: 
     'CLVTools.R'
     'RcppExports.R'
     'all_generics.R'
+    'catch-routine-registration.R'
     'class_clv_time.R'
     'class_clv_data.R'
     'class_clv_model.R'
@@ -112,12 +114,14 @@ Collate:
     'f_interface_clvdata.R'
     'f_interface_gg.R'
     'f_interface_ggomnbd.R'
+    'f_interface_latentattrition.R'
     'f_interface_pmf.R'
     'f_interface_pnbd.R'
     'f_interface_predict_clvfittedspending.R'
     'f_interface_predict_clvfittedtransactions.R'
     'f_interface_setdynamiccovariates.R'
     'f_interface_setstaticcovariates.R'
+    'f_interface_spending.R'
     'f_s3generics_clvdata.R'
     'f_s3generics_clvdata_dynamiccov.R'
     'f_s3generics_clvdata_plot.R'
@@ -137,10 +141,10 @@ Collate:
     'pnbd_dyncov_BkSum.R'
     'pnbd_dyncov_CET.R'
     'pnbd_dyncov_DECT.R'
-    'pnbd_dyncov_LL.R'
     'pnbd_dyncov_createwalks.R'
     'pnbd_dyncov_expectation.R'
-    'pnbd_dyncov_makewalks.R'
     'pnbd_dyncov_palive.R'
-RoxygenNote: 7.1.2
+RoxygenNote: 7.2.3
 VignetteBuilder: knitr
+Config/testthat/parallel: false
+Config/testthat/edition: 3
diff --git a/NAMESPACE b/NAMESPACE
@@ -41,6 +41,8 @@ export(SetDynamicCovariates)
 export(SetStaticCovariates)
 export(as.clv.data)
 export(clvdata)
+export(latentAttrition)
+export(spending)
 exportMethods(bgbb)
 exportMethods(bgnbd)
 exportMethods(gg)
@@ -52,30 +54,38 @@ exportMethods(predict)
 exportMethods(show)
 import(data.table)
 import(optimx)
+importFrom(Formula,as.Formula)
+importFrom(Formula,is.Formula)
 importFrom(MASS,ginv)
 importFrom(Matrix,nearPD)
 importFrom(Rcpp,evalCpp)
 importFrom(ggplot2,aes)
-importFrom(ggplot2,aes_string)
 importFrom(ggplot2,element_blank)
 importFrom(ggplot2,element_line)
 importFrom(ggplot2,element_rect)
 importFrom(ggplot2,element_text)
 importFrom(ggplot2,geom_col)
 importFrom(ggplot2,geom_line)
+importFrom(ggplot2,geom_point)
+importFrom(ggplot2,geom_segment)
 importFrom(ggplot2,geom_text)
 importFrom(ggplot2,geom_vline)
 importFrom(ggplot2,ggplot)
+importFrom(ggplot2,ggtitle)
 importFrom(ggplot2,guide_legend)
 importFrom(ggplot2,labs)
 importFrom(ggplot2,position_dodge)
 importFrom(ggplot2,position_dodge2)
 importFrom(ggplot2,rel)
 importFrom(ggplot2,scale_colour_manual)
 importFrom(ggplot2,scale_fill_manual)
+importFrom(ggplot2,scale_x_date)
+importFrom(ggplot2,scale_x_datetime)
 importFrom(ggplot2,scale_x_discrete)
 importFrom(ggplot2,stat_density)
 importFrom(ggplot2,theme)
+importFrom(ggplot2,xlab)
+importFrom(ggplot2,ylim)
 importFrom(graphics,plot)
 importFrom(lubridate,ceiling_date)
 importFrom(lubridate,days)
@@ -105,6 +115,7 @@ importFrom(stats,AIC)
 importFrom(stats,BIC)
 importFrom(stats,coef)
 importFrom(stats,fitted)
+importFrom(stats,formula)
 importFrom(stats,integrate)
 importFrom(stats,logLik)
 importFrom(stats,model.frame)
@@ -118,6 +129,8 @@ importFrom(stats,qnorm)
 importFrom(stats,reformulate)
 importFrom(stats,sd)
 importFrom(stats,setNames)
+importFrom(stats,terms)
+importFrom(stats,update)
 importFrom(stats,vcov)
 importFrom(utils,getFromNamespace)
 importFrom(utils,modifyList)

diff --git a/NEWS.md b/NEWS.md
@@ -1,3 +1,13 @@
+# CLVTools 0.10.0
+
+### NEW FEATURES
+* We add an interface to specify models using a formula notation (`latentAttrition()` and `spending()`)
+* New method to plot customer's transaction timings (`plot.clv.data(which='timings')`)
+* Draw diagnostic plots of multiple models in single plot (`plot(other.models=list(), label=c())`)
+* MUCH faster fitting for the Pareto/NBD with time-varying covariates because we implemented the LL in Rcpp
+
+
+
 # CLVTools 0.9.0
 
 ### NEW FEATURES

diff --git a/R/RcppExports.R b/R/RcppExports.R
@@ -368,7 +368,7 @@ NULL
 #' @description
 #' Calculates the discounted expected residual transactions.
 #'
-#' \itemize{
+#' \describe{
 #' \item{\code{pnbd_nocov_DERT}}{ Discounted expected residual transactions for the Pareto/NBD model without covariates}
 #' \item{\code{pnbd_staticcov_DERT}}{ Discounted expected residual transactions for the Pareto/NBD model with static covariates}
 #' }
@@ -534,23 +534,11 @@ pnbd_staticcov_PMF <- function(r, s, x, vAlpha_i, vBeta_i, vT_i) {
     .Call(`_CLVTools_pnbd_staticcov_PMF`, r, s, x, vAlpha_i, vBeta_i, vT_i)
 }
 
-pnbd_dyncov_LL_Bi_cpp <- function(i, t_x, d, delta, n_walks, max_walks, walks) {
-    .Call(`_CLVTools_pnbd_dyncov_LL_Bi_cpp`, i, t_x, d, delta, n_walks, max_walks, walks)
+pnbd_dyncov_LL_negsum <- function(params, X, t_x, T_cal, d_omega, walkinfo_aux_life, walkinfo_real_life, walkinfo_aux_trans, walkinfo_real_trans, walkinfo_trans_real_from, walkinfo_trans_real_to, covdata_aux_life, covdata_real_life, covdata_aux_trans, covdata_real_trans) {
+    .Call(`_CLVTools_pnbd_dyncov_LL_negsum`, params, X, t_x, T_cal, d_omega, walkinfo_aux_life, walkinfo_real_life, walkinfo_aux_trans, walkinfo_real_trans, walkinfo_trans_real_from, walkinfo_trans_real_to, covdata_aux_life, covdata_real_life, covdata_aux_trans, covdata_real_trans)
 }
 
-pnbd_dyncov_LL_Di_cpp <- function(i, real_d, aux_d, real_n_walks, aux_n_walks, real_max_walks, aux_max_walks, real_adj_walk1, real_walks, aux_walks) {
-    .Call(`_CLVTools_pnbd_dyncov_LL_Di_cpp`, i, real_d, aux_d, real_n_walks, aux_n_walks, real_max_walks, aux_max_walks, real_adj_walk1, real_walks, aux_walks)
-}
-
-hyp_alpha_ge_beta_cpp <- function(alpha_1, beta_1, alpha_2, beta_2, x, r, s) {
-    .Call(`_CLVTools_hyp_alpha_ge_beta_cpp`, alpha_1, beta_1, alpha_2, beta_2, x, r, s)
-}
-
-hyp_beta_g_alpha_cpp <- function(alpha_1, beta_1, alpha_2, beta_2, x, r, s) {
-    .Call(`_CLVTools_hyp_beta_g_alpha_cpp`, alpha_1, beta_1, alpha_2, beta_2, x, r, s)
-}
-
-F2_3_vecs_cpp <- function(n_walks_cbs, dT_cbs, Bjsum_cbs, x_cbs, t_x_cbs, n_walks_trans, walks_trans, d_trans, delta_trans, max_walks_trans, n_walks_life_real, d_life_real, max_walks_life_real, adj_walk1_life_real, walks_life_real, n_walks_life_aux, d_life_aux, max_walks_life_aux, walks_life_aux, r, alpha, s, beta) {
-    .Call(`_CLVTools_F2_3_vecs_cpp`, n_walks_cbs, dT_cbs, Bjsum_cbs, x_cbs, t_x_cbs, n_walks_trans, walks_trans, d_trans, delta_trans, max_walks_trans, n_walks_life_real, d_life_real, max_walks_life_real, adj_walk1_life_real, walks_life_real, n_walks_life_aux, d_life_aux, max_walks_life_aux, walks_life_aux, r, alpha, s, beta)
+pnbd_dyncov_LL_ind <- function(params, X, t_x, T_cal, d_omega, walkinfo_aux_life, walkinfo_real_life, walkinfo_aux_trans, walkinfo_real_trans, walkinfo_trans_real_from, walkinfo_trans_real_to, covdata_aux_life, covdata_real_life, covdata_aux_trans, covdata_real_trans, return_intermediate_results = FALSE) {
+    .Call(`_CLVTools_pnbd_dyncov_LL_ind`, params, X, t_x, T_cal, d_omega, walkinfo_aux_life, walkinfo_real_life, walkinfo_aux_trans, walkinfo_real_trans, walkinfo_trans_real_from, walkinfo_trans_real_to, covdata_aux_life, covdata_real_life, covdata_aux_trans, covdata_real_trans, return_intermediate_results)
 }
 
diff --git a/R/all_generics.R b/R/all_generics.R
@@ -206,11 +206,13 @@ setGeneric("clv.time.format.timepoint", function(clv.time, timepoint)
 #' See section "Details" of \link{clvdata} for more details on parameters and usage.
 #'
 #' @examples
+#' \donttest{ # dont test because ncpu=2 limit on cran (too fast)
 #' data(cdnow)
 #'
 #' # Turn data.table of transaction data into a clv.data object,
 #' #  using default date format and column names but no holdout period
 #' clv.cdnow <- as.clv.data(cdnow)
+#' }
 #'
 #' @export
 as.clv.data <- function(x,

diff --git a/R/catch-routine-registration.R b/R/catch-routine-registration.R
@@ -0,0 +1,6 @@
+# This dummy function definition is included with the package to ensure that
+# 'tools::package_native_routine_registration_skeleton()' generates the required
+# registration info for the 'run_testthat_tests' symbol.
+(function() {
+  .Call("run_testthat_tests", FALSE, PACKAGE = "CLVTools")
+})
diff --git a/R/class_clv_model_pnbd.R b/R/class_clv_model_pnbd.R
@@ -13,7 +13,7 @@ setClass(Class = "clv.model.pnbd.no.cov", contains = "clv.model.with.correlation
 clv.model.pnbd.no.cov <- function(){
 
   return(new("clv.model.pnbd.no.cov",
-             name.model                  = "Pareto NBD Standard",
+             name.model                  = "Pareto/NBD Standard",
              names.original.params.model = c(r="r", alpha="alpha", s="s", beta="beta"),
              names.prefixed.params.model = c("log.r","log.alpha", "log.s", "log.beta"),
              start.params.model          = c(r=1, alpha=1, s=1, beta=1),