Construct the full ROC curve given probability estimates and true classifications.
Arguments
- estimates
A vector of classification probabilities. Values should represent the probability of
1
in thetruth
argument.- truth
An integer vector of
0
and1
representing the true classifications.
Value
An roc_df object.
Examples
create_roc(estimates = dcm_probs$att1$estimate, truth = dcm_probs$att1$truth)
#> # A tibble: 418 × 3
#> .threshold specificity sensitivity
#> <dbl> <dbl> <dbl>
#> 1 -Inf 0 1
#> 2 0.0000716 0 1
#> 3 0.000336 0.00316 1
#> 4 0.000390 0.00633 1
#> 5 0.000684 0.00949 1
#> 6 0.000694 0.0127 1
#> 7 0.000808 0.0158 1
#> 8 0.00109 0.0190 1
#> 9 0.00111 0.0222 1
#> 10 0.00122 0.0253 1
#> # ℹ 408 more rows