diff --git a/bob/ip/binseg/utils/measure.py b/bob/ip/binseg/utils/measure.py index 35bbd18fc3526714871df27329c857fd4c037215..07a95a1414b9eafc361a75f091beee72ba9ad8ed 100644 --- a/bob/ip/binseg/utils/measure.py +++ b/bob/ip/binseg/utils/measure.py @@ -29,55 +29,51 @@ class SmoothedValue: def base_measures(tp, fp, tn, fn): - """ - Calculates a bunch of measures from true/false positive and negative counts + """Calculates measures from true/false positive and negative counts This function can return standard machine learning measures from true and false positive counts of positives and negatives. For a thorough look into these and alternate names for the returned values, please check Wikipedia's - entry on `Precision and Recall`_. + entry on `Precision and Recall + <https://en.wikipedia.org/wiki/Precision_and_recall>`_. Parameters ---------- - tp : int - True positive count, AKA "hit" + tp : int + True positive count, AKA "hit" - fp : int - False positive count, AKA, "correct rejection" + fp : int + False positive count, AKA, "correct rejection" - tn : int - True negative count, AKA "false alarm", or "Type I error" + tn : int + True negative count, AKA "false alarm", or "Type I error" - fn : int - False Negative count, AKA "miss", or "Type II error" + fn : int + False Negative count, AKA "miss", or "Type II error" Returns ------- - precision : float - P, AKA positive predictive value (PPV) - :math:`\frac{tp}{tp+fp}` + precision : float + P, AKA positive predictive value (PPV). - recall : float - R, AKA sensitivity, hit rate, or true positive rate (TPR) - :math:`\frac{tp}{p} = \frac{tp}{tp+fn}` + recall : float + R, AKA sensitivity, hit rate, or true positive rate (TPR). - specificity : float - S, AKA selectivity or true negative rate (TNR). - :math:`\frac{tn}{n} = \frac{tn}{tn+fp}` + specificity : float + S, AKA selectivity or true negative rate (TNR). - accuracy : float - A, :math:`\frac{tp + tn}{p + n} = \frac{tp + tn}{tp + fp + tn + fn}` + accuracy : float + A - jaccard : float - J, :math:`\frac{tp}{tp+fp+fn}`, see `Jaccard Index`_ + jaccard : float + J, see `Jaccard Index <https://en.wikipedia.org/wiki/Jaccard_index>`_ - f1_score : float - F1, :math:`\frac{2 P R}{P + R} = \frac{2tp}{2tp + fp + fn}`, see - `F1-score`_ + f1_score : float + F1, see `F1-score <https://en.wikipedia.org/wiki/F1_score>`_ """ diff --git a/doc/api.rst b/doc/api.rst index 112d60a2bb686bb5456445f08fb2b3e3abff36ed..f1e5cd3c7efc795fd2e931be975c6e27601fd768 100644 --- a/doc/api.rst +++ b/doc/api.rst @@ -185,3 +185,6 @@ Datasets bob.ip.binseg.configs.datasets.drionsdb.expert1 bob.ip.binseg.configs.datasets.drionsdb.expert2 + + +.. include:: links.rst diff --git a/doc/links.rst b/doc/links.rst index 5d0ffb981bf0265159e33e4a45a8ca3326422c0e..ce2c72d68eb7f9a94b4704b39e30cbb3da111552 100644 --- a/doc/links.rst +++ b/doc/links.rst @@ -9,9 +9,6 @@ .. _pytorch: https://pytorch.org .. _tabulate: https://pypi.org/project/tabulate/ .. _our paper: https://arxiv.org/abs/1909.03856 -.. _precision and recall: https://en.wikipedia.org/wiki/Precision_and_recall -.. _f1-score: https://en.wikipedia.org/wiki/F1_score -.. _jaccard index: https://en.wikipedia.org/wiki/Jaccard_index .. Raw data websites .. _drive: https://www.isi.uu.nl/Research/Databases/DRIVE/