declearn.metrics.BinaryAccuracyPrecisionRecall
Bases: Metric[BinaryConfmat]
Binary classification accuracy, precision and recall metrics.
This metric applies to binary classifier, and computes the (opt. weighted) amount of true positives (TP), true negatives (TN), false positives (FP) and false negatives (FN) predictions over time, from which basic evaluation metrics may be derived.
Computed metrics are the following:
- accuracy: float Accuracy of the classifier, i.e. P(pred==true). Formula: (TP + TN) / (TP + TN + FP + FN)
- precision: float Precision score, i.e. P(true=1|pred=1). Formula: TP / (TP + FP)
- recall: float Recall score, i.e. P(pred=1|true=1). Formula: TP / (TP + FN)
- f-score: float F1-score, i.e. harmonic mean of precision and recall. Formula: (2TP) / (2TP + FP + FN)
- confusion: 2-d numpy.ndarray Confusion matrix of predictions. Values: [[TN, FP], [FN, TP]]
Source code in declearn/metrics/_classif.py
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 |
|
__init__(thresh=0.5, label=1)
Instantiate the binary accuracy / precision / recall metrics.
Parameters:
Name | Type | Description | Default |
---|---|---|---|
thresh |
float
|
Threshold value around which to binarize input predictions. |
0.5
|
label |
Union[int, str]
|
Value of the positive labels in input true-label arrays. |
1
|
Source code in declearn/metrics/_classif.py
75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 |
|