Merge branch 'main' into release-news-update

Author: andrewherren

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: stochtree
 Title: Stochastic Tree Ensembles (XBART and BART) for Supervised Learning and Causal Inference
-Version: 0.1.1
+Version: 0.2.0
 Authors@R: 
   c(
     person("Drew", "Herren", email = "drewherrenopensource@gmail.com", role = c("aut", "cre"), comment = c(ORCID = "0000-0003-4109-6611")), 

Doxyfile

@@ -48,7 +48,7 @@ PROJECT_NAME           = "StochTree"
 # could be handy for archiving the generated documentation or if some version
 # control system is used.
 
-PROJECT_NUMBER         = 0.1.1
+PROJECT_NUMBER         = 0.2.0
 
 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer a

R/bart.R

@@ -423,37 +423,53 @@ bart <- function(
       floor(num_values / cutpoint_grid_size),
       1
     )
+    x_is_df <- is.data.frame(X_train)
     covs_warning_1 <- NULL
     covs_warning_2 <- NULL
     covs_warning_3 <- NULL
+    covs_warning_4 <- NULL
     for (i in 1:num_cov_orig) {
-      # Determine the number of unique values
-      num_unique_values <- length(unique(X_train[, i]))
-
-      # Determine a "name" for the covariate
-      cov_name <- ifelse(
-        is.null(colnames(X_train)),
-        paste0("X", i),
-        colnames(X_train)[i]
-      )
-
-      # Check for a small relative number of unique values
-      unique_full_ratio <- num_unique_values / num_values
-      if (unique_full_ratio < 0.2) {
-        covs_warning_1 <- c(covs_warning_1, cov_name)
+      # Skip check for variables that are treated as categorical
+      x_numeric <- T
+      if (x_is_df) {
+        if (is.factor(X_train[, i])) {
+          x_numeric <- F
+        }
       }
+      if (x_numeric) {
+        # Determine the number of unique values
+        num_unique_values <- length(unique(X_train[, i]))
+
+        # Determine a "name" for the covariate
+        cov_name <- ifelse(
+          is.null(colnames(X_train)),
+          paste0("X", i),
+          colnames(X_train)[i]
+        )
 
-      # Check for a small absolute number of unique values
-      if (num_values > 100) {
-        if (num_unique_values < 20) {
-          covs_warning_2 <- c(covs_warning_2, cov_name)
+        # Check for a small relative number of unique values
+        unique_full_ratio <- num_unique_values / num_values
+        if (unique_full_ratio < 0.2) {
+          covs_warning_1 <- c(covs_warning_1, cov_name)
+        }
+
+        # Check for a small absolute number of unique values
+        if (num_values > 100) {
+          if (num_unique_values < 20) {
+            covs_warning_2 <- c(covs_warning_2, cov_name)
+          }
+        }
+
+        # Check for a large number of duplicates of any individual value
+        x_j_hist <- table(X_train[, i])
+        if (any(x_j_hist > 2 * max_grid_size)) {
+          covs_warning_3 <- c(covs_warning_3, cov_name)
         }
-      }
 
-      # Check for a large number of duplicates of any individual value
-      x_j_hist <- table(X_train[, i])
-      if (any(x_j_hist > 2 * max_grid_size)) {
-        covs_warning_3 <- c(covs_warning_3, cov_name)
+        # Check for binary variables
+        if (num_unique_values == 2) {
+          covs_warning_4 <- c(covs_warning_4, cov_name)
+        }
       }
     }
 
@@ -494,6 +510,18 @@ bart <- function(
         )
       )
     }
+
+    if (!is.null(covs_warning_4)) {
+      warning(
+        paste0(
+          "Covariates ",
+          paste(covs_warning_4, collapse = ", "),
+          " appear to be binary but are currently treated by stochtree as continuous. ",
+          "This might present some issues with the grow-from-root (GFR) algorithm. ",
+          "Consider converting binary variables to ordered factor (i.e. `factor(..., ordered = T)`."
+        )
+      )
+    }
   }
 
   # Standardize the keep variable lists to numeric indices

R/bcf.R

@@ -527,37 +527,54 @@ bcf <- function(
       floor(num_values / cutpoint_grid_size),
       1
     )
+    x_is_df <- is.data.frame(X_train)
     covs_warning_1 <- NULL
     covs_warning_2 <- NULL
     covs_warning_3 <- NULL
+    covs_warning_4 <- NULL
     for (i in 1:num_cov_orig) {
-      # Determine the number of unique values
-      num_unique_values <- length(unique(X_train[, i]))
-
-      # Determine a "name" for the covariate
-      cov_name <- ifelse(
-        is.null(colnames(X_train)),
-        paste0("X", i),
-        colnames(X_train)[i]
-      )
-
-      # Check for a small relative number of unique values
-      unique_full_ratio <- num_unique_values / num_values
-      if (unique_full_ratio < 0.2) {
-        covs_warning_1 <- c(covs_warning_1, cov_name)
+      # Skip check for variables that are treated as categorical
+      x_numeric <- T
+      if (x_is_df) {
+        if (is.factor(X_train[, i])) {
+          x_numeric <- F
+        }
       }
 
-      # Check for a small absolute number of unique values
-      if (num_values > 100) {
-        if (num_unique_values < 20) {
-          covs_warning_2 <- c(covs_warning_2, cov_name)
+      if (x_numeric) {
+        # Determine the number of unique values
+        num_unique_values <- length(unique(X_train[, i]))
+
+        # Determine a "name" for the covariate
+        cov_name <- ifelse(
+          is.null(colnames(X_train)),
+          paste0("X", i),
+          colnames(X_train)[i]
+        )
+
+        # Check for a small relative number of unique values
+        unique_full_ratio <- num_unique_values / num_values
+        if (unique_full_ratio < 0.2) {
+          covs_warning_1 <- c(covs_warning_1, cov_name)
         }
-      }
 
-      # Check for a large number of duplicates of any individual value
-      x_j_hist <- table(X_train[, i])
-      if (any(x_j_hist > 2 * max_grid_size)) {
-        covs_warning_3 <- c(covs_warning_3, cov_name)
+        # Check for a small absolute number of unique values
+        if (num_values > 100) {
+          if (num_unique_values < 20) {
+            covs_warning_2 <- c(covs_warning_2, cov_name)
+          }
+        }
+
+        # Check for a large number of duplicates of any individual value
+        x_j_hist <- table(X_train[, i])
+        if (any(x_j_hist > 2 * max_grid_size)) {
+          covs_warning_3 <- c(covs_warning_3, cov_name)
+        }
+
+        # Check for binary variables
+        if (num_unique_values == 2) {
+          covs_warning_4 <- c(covs_warning_4, cov_name)
+        }
       }
     }
 
@@ -598,6 +615,18 @@ bcf <- function(
         )
       )
     }
+
+    if (!is.null(covs_warning_4)) {
+      warning(
+        paste0(
+          "Covariates ",
+          paste(covs_warning_4, collapse = ", "),
+          " appear to be binary but are currently treated by stochtree as continuous. ",
+          "This might present some issues with the grow-from-root (GFR) algorithm. ",
+          "Consider converting binary variables to ordered factor (i.e. `factor(..., ordered = T)`."
+        )
+      )
+    }
   }
 
   # Check delta_max is valid

cleanup

@@ -0,0 +1,2 @@
+#!/bin/sh
+rm -f src/Makevars

configure

@@ -1,6 +1,6 @@
 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.72 for stochtree 0.1.1.
+# Generated by GNU Autoconf 2.72 for stochtree 0.2.0.
 #
 #
 # Copyright (C) 1992-1996, 1998-2017, 2020-2023 Free Software Foundation,
@@ -600,8 +600,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='stochtree'
 PACKAGE_TARNAME='stochtree'
-PACKAGE_VERSION='0.1.1'
-PACKAGE_STRING='stochtree 0.1.1'
+PACKAGE_VERSION='0.2.0'
+PACKAGE_STRING='stochtree 0.2.0'
 PACKAGE_BUGREPORT=''
 PACKAGE_URL=''
 
@@ -1205,7 +1205,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-'configure' configures stochtree 0.1.1 to adapt to many kinds of systems.
+'configure' configures stochtree 0.2.0 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1267,7 +1267,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of stochtree 0.1.1:";;
+     short | recursive ) echo "Configuration of stochtree 0.2.0:";;
    esac
   cat <<\_ACEOF
 
@@ -1335,7 +1335,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-stochtree configure 0.1.1
+stochtree configure 0.2.0
 generated by GNU Autoconf 2.72
 
 Copyright (C) 2023 Free Software Foundation, Inc.
@@ -1372,7 +1372,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by stochtree $as_me 0.1.1, which was
+It was created by stochtree $as_me 0.2.0, which was
 generated by GNU Autoconf 2.72.  Invocation command line was
 
   $ $0$ac_configure_args_raw
@@ -2385,7 +2385,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by stochtree $as_me 0.1.1, which was
+This file was extended by stochtree $as_me 0.2.0, which was
 generated by GNU Autoconf 2.72.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -2440,7 +2440,7 @@ ac_cs_config_escaped=`printf "%s\n" "$ac_cs_config" | sed "s/^ //; s/'/'\\\\\\\\
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config='$ac_cs_config_escaped'
 ac_cs_version="\\
-stochtree config.status 0.1.1
+stochtree config.status 0.2.0
 configured by $0, generated by GNU Autoconf 2.72,
   with options \\"\$ac_cs_config\\"
 
@@ -3000,3 +3000,4 @@ if test -n "$ac_unrecognized_opts" && test "$enable_option_checking" != no; then
 printf "%s\n" "$as_me: WARNING: unrecognized options: $ac_unrecognized_opts" >&2;}
 fi
 
+

configure.ac

@@ -3,7 +3,7 @@
 # https://github.com/microsoft/LightGBM/blob/master/R-package/configure.ac
 
 AC_PREREQ(2.69)
-AC_INIT([stochtree], [0.1.1], [], [stochtree], [])
+AC_INIT([stochtree], [0.2.0], [], [stochtree], [])
 # Note: consider making version number dynamic as in 
 # https://github.com/microsoft/LightGBM/blob/195c26fc7b00eb0fec252dfe841e2e66d6833954/build-cran-package.sh