analytics~metrics. ClassificationScore
Source: analyticsdoc.
Class implements several classification measures (precision, recall, F1, accuracy).
Property
Method
new ClassificationScore(yTrue, yPred)
For evaluating provided categories from binary? classifiers.
Parameters
| Name | Type | Optional | Description |
|---|---|---|---|
|
yTrue |
(Array of number or module:la.Vector) |
|
Ground truth (correct) lable(s). |
|
yPred |
(Array of number or module:la.Vector) |
|
Predicted (estimated) lable(s). |
Property
scores
Returns Object containing different classification measures.
- Returns
-
ObjectB scores - Object with different classification socres. -
numberB scores.count - Count. -
numberB scores.TP - Number of true positives. -
numberB scores.TN - Number of true negative. -
numberB scores.FP - Number of false positives. -
numberB scores.FN - Number of false positives. -
numberB scores.all - Number of all results. -
numberB scores.accuracy - Accuracy score. Formula:(tp + tn) / (tp + fp + fn + tn). -
numberB scores.precision - Precision score. Formula:tp / (tp + fp). -
numberB scores.recall - Recall score. Formula:tp / (tp + fn). -
numberB scores.f1 - F1 score. Formula:2 * (precision * recall) / (precision + recall).
Method
push(correct, predicted)
Adds prediction to the current statistics. Labels can be either integers. or integer array (when there are zero or more then one lables).
Parameters
| Name | Type | Optional | Description |
|---|---|---|---|
|
correct |
number |
|
Correct lable. |
|
predicted |
number |
|
Predicted lable. |