Merge branch 'main' of https://github.com/SchlossLab/mikropml

SchlossLab · Feb 16, 2022 · 60608a5 · 60608a5
2 parents 76a019d + 6572f78
commit 60608a5
Show file tree

Hide file tree

Showing 23 changed files with 36 additions and 27 deletions.
diff --git a/NEWS.md b/NEWS.md
@@ -1,6 +1,7 @@
 # development version
 
 - mikropml now requires R version 4.1.0 or greater due to an update in the randomForest package (#292). 
+- Fix bug where `cv_times` had no effect on repeats for cross-validation (#291, @kelly-sovacool).
 
 # mikropml 1.2.2
 

diff --git a/R/cross_val.R b/R/cross_val.R
@@ -47,6 +47,7 @@ define_cv <- function(train_data, outcome_colname, hyperparams_list, perf_metric
   cv <- caret::trainControl(
     method = "repeatedcv",
     number = kfold,
+    repeats = cv_times,
     index = cvIndex,
     returnResamp = "final",
     classProbs = class_probs,

diff --git a/data-raw/otu_mini_bin.R b/data-raw/otu_mini_bin.R
@@ -63,7 +63,7 @@ otu_mini_cv <- define_cv(otu_mini_bin_results_glmnet$trained_model$trainingData,
   perf_metric_function = caret::multiClassSummary,
   class_probs = TRUE,
   cv_times = 2,
-  group = cv_group
+  groups = cv_group
 )
 usethis::use_data(otu_mini_cv, overwrite = TRUE)
 
@@ -87,7 +87,7 @@ otu_mini_bin_results_svmRadial <- mikropml::run_ml(otu_mini_bin,
 )
 usethis::use_data(otu_mini_bin_results_svmRadial, overwrite = TRUE)
 
-otu_mini_bin_results_xgbTree <- mikropml::run_ml(otu_mini,
+otu_mini_bin_results_xgbTree <- mikropml::run_ml(otu_mini_bin,
   "xgbTree",
   outcome_colname = "dx",
   find_feature_importance = FALSE,

diff --git a/data/otu_mini_bin.rda b/data/otu_mini_bin.rda
diff --git a/data/otu_mini_bin_results_glmnet.rda b/data/otu_mini_bin_results_glmnet.rda
diff --git a/data/otu_mini_bin_results_rf.rda b/data/otu_mini_bin_results_rf.rda
diff --git a/data/otu_mini_bin_results_rpart2.rda b/data/otu_mini_bin_results_rpart2.rda
diff --git a/data/otu_mini_bin_results_svmRadial.rda b/data/otu_mini_bin_results_svmRadial.rda
diff --git a/data/otu_mini_bin_results_xgbTree.rda b/data/otu_mini_bin_results_xgbTree.rda
diff --git a/data/otu_mini_cont_results_glmnet.rda b/data/otu_mini_cont_results_glmnet.rda
diff --git a/data/otu_mini_cont_results_nocv.rda b/data/otu_mini_cont_results_nocv.rda
diff --git a/data/otu_mini_cv.rda b/data/otu_mini_cv.rda
diff --git a/data/otu_mini_multi.rda b/data/otu_mini_multi.rda
diff --git a/data/otu_mini_multi_group.rda b/data/otu_mini_multi_group.rda
diff --git a/data/otu_mini_multi_results_glmnet.rda b/data/otu_mini_multi_results_glmnet.rda
diff --git a/data/otu_small.rda b/data/otu_small.rda
diff --git a/docs/articles/introduction.html b/docs/articles/introduction.html
diff --git a/docs/articles/tuning.html b/docs/articles/tuning.html
diff --git a/docs/pkgdown.yml b/docs/pkgdown.yml
@@ -7,7 +7,7 @@ articles:
   parallel: parallel.html
   preprocess: preprocess.html
   tuning: tuning.html
-last_built: 2022-02-16T00:57Z
+last_built: 2022-02-16T16:38Z
 urls:
   reference: http://www.schlosslab.org/mikropml/reference
   article: http://www.schlosslab.org/mikropml/articles

diff --git a/docs/reference/get_perf_metric_fn.html b/docs/reference/get_perf_metric_fn.html
diff --git a/docs/reference/randomize_feature_order.html b/docs/reference/randomize_feature_order.html
diff --git a/tests/testthat/test-cross_val.R b/tests/testthat/test-cross_val.R
@@ -21,20 +21,22 @@ cv_group <- c(
 
 test_that("define_cv works on otu_mini training data with groups", {
   set.seed(2019)
-  expect_equal(
-    expect_message(
-      define_cv(otu_mini_bin_results_rf$trained_model$trainingData,
-        ".outcome",
-        hparams_list,
-        perf_metric_function,
-        class_probs = class_probs,
-        cv_times = 2,
-        groups = cv_group
-      ),
-      "Groups will be kept together in CV partitions"
+  expect_message(
+    cv <- define_cv(otu_mini_bin_results_rf$trained_model$trainingData,
+      ".outcome",
+      hparams_list,
+      perf_metric_function,
+      class_probs = class_probs,
+      cv_times = 2,
+      groups = cv_group
     ),
-    otu_mini_cv
+    "Groups will be kept together in CV partitions"
   )
+  expect_equal(cv, otu_mini_cv)
+  expect_equal(cv$method, "repeatedcv")
+  expect_equal(cv$number, 5)
+  expect_equal(cv$repeats, 2)
+
   expect_message(
     define_cv(otu_mini_bin_results_rf$trained_model$trainingData,
       ".outcome",

diff --git a/tests/testthat/test-run_ml.R b/tests/testthat/test-run_ml.R
@@ -251,3 +251,8 @@ test_that("run_ml catches bad training_frac values", {
     "`training_frac` must be a numeric between 0 and 1."
   )
 })
+
+test_that("models use repeatedcv", {
+  expect_equal(otu_mini_bin_results_glmnet$trained_model$control$method, "repeatedcv")
+  expect_equal(otu_mini_bin_results_glmnet$trained_model$control$repeats, 2)
+})