| Title: | Aggregate Multiple ROC Curves into One Global ROC |
|---|---|
| Description: | Aggregates multiple Receiver Operating Characteristic (ROC) curves obtained from different sources into one global ROC. Additionally, it’s also possible to calculate the aggregated precision-recall (PR) curve. |
| Authors: | Pedro Mateus [aut, cre]
|
| Maintainer: | Pedro Mateus <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 1.0.1 |
| Built: | 2024-12-04 07:09:41 UTC |
| Source: | CRAN |
Compute the global confusion matrix from the FPR and TPR obtained from each node
partial_cm( fpr, tpr, thresholds, negative_count, total_count, descending = FALSE )partial_cm( fpr, tpr, thresholds, negative_count, total_count, descending = FALSE )
fpr |
list - False positive rates for each individual ROC |
tpr |
list - True positive rates for each individual ROC |
thresholds |
list - Thresholds used to compute the fpr and tpr |
negative_count |
list - Total number of samples corresponding to the negative case |
total_count |
list - Total number of samples |
descending |
thresholds in descending order? |
global confusion matrix and thresholds
Compute the precision recall curve
precision_recall_curve(fpr, tpr, thresholds, negative_count, total_count)precision_recall_curve(fpr, tpr, thresholds, negative_count, total_count)
fpr |
list - False positive rates for each individual ROC. |
tpr |
list - True positive rates for each individual ROC. |
thresholds |
list - Thresholds used to compute the fpr and tpr. |
negative_count |
vector - Total number of samples corresponding to the negative case. |
total_count |
vector - Total number of samples. |
list with the global precision, recall, and thresholds (increasing)
Compute Receiver operating characteristic (ROC)
roc_curve(fpr, tpr, thresholds, negative_count, total_count)roc_curve(fpr, tpr, thresholds, negative_count, total_count)
fpr |
list - False positive rates for each individual ROC |
tpr |
list - True positive rates for each individual ROC |
thresholds |
list - Thresholds used to compute the fpr and tpr |
negative_count |
vector - Total number of samples corresponding to the negative case |
total_count |
vector - Total number of samples |
list with the global fpr, tpr, and thresholds (decreasing)
Shift a vector left or right according to the value provided
shift_vector(x, n)shift_vector(x, n)
x |
the vector |
n |
shift |
the vector shifted
shift_vector(c(1,2,3,4), 1) shift_vector(c(1,2,3,4), -1)shift_vector(c(1,2,3,4), 1) shift_vector(c(1,2,3,4), -1)