Calculate the fraction of positives, i.e. baseline precision for a PRC curve
+ Source:R/performance.R
+ calc_baseline_precision.Rd
Calculate the fraction of positives, i.e. baseline precision for a PRC curve
+Arguments
+- dataset +
Dataframe with an outcome variable and other columns as features.
+
+
+- outcome_colname +
Column name as a string of the outcome variable +(default
NULL
; the first column will be chosen automatically).
+
+
+- pos_outcome +
the positive outcome from
outcome_colname
, +e.g. "cancer" for theotu_mini_bin
dataset.
+
+
Author
+Kelly Sovacool, sovacool@umich.edu
+Examples
+# calculate the baseline precision
+data.frame(y = c("a", "b", "a", "b")) %>%
+ calc_baseline_precision("y", "a")
+#> Using 'y' as the outcome column.
+#> [1] 0.5
+
+
+calc_baseline_precision(otu_mini_bin,
+ outcome_colname = "dx",
+ pos_outcome = "cancer")
+#> Using 'dx' as the outcome column.
+#> [1] 0.49
+
+
+# if you're not sure which outcome was used as the 'positive' outcome during
+# model training, you can access it from the trained model and pass it along:
+calc_baseline_precision(otu_mini_bin,
+ outcome_colname = "dx",
+ pos_outcome = otu_mini_bin_results_glmnet$trained_model$levels[1])
+#> Using 'dx' as the outcome column.
+#> [1] 0.49
+
+
+