leslyarun commited on
Commit
4b1329a
·
1 Parent(s): 1a601d7

Update fbeta_score.py

Browse files
Files changed (1) hide show
  1. fbeta_score.py +11 -12
fbeta_score.py CHANGED
@@ -46,6 +46,7 @@ _KWARGS_DESCRIPTION = """
46
  Args:
47
  predictions (`list` of `int`): Predicted labels.
48
  references (`list` of `int`): Ground truth labels.
 
49
  labels (`list` of `int`): The set of labels to include when `average` is not set to `'binary'`, and the order of the labels if `average` is `None`. Labels present in the data can be excluded, for example to calculate a multiclass average ignoring a majority negative class. Labels not present in the data will result in 0 components in a macro average. For multilabel targets, labels are column indices. By default, all labels in `predictions` and `references` are used in sorted order. Defaults to None.
50
  pos_label (`int`): The class to be considered the positive class, in the case where `average` is set to `binary`. Defaults to 1.
51
  average (`string`): This parameter is required for multiclass/multilabel targets. If set to `None`, the scores for each class are returned. Otherwise, this determines the type of averaging performed on the data. Defaults to `'binary'`.
@@ -55,20 +56,18 @@ Args:
55
  - 'weighted': Calculate metrics for each label, and find their average weighted by support (the number of true instances for each label). This alters `'macro'` to account for label imbalance. This option can result in an F-score that is not between precision and recall.
56
  - 'samples': Calculate metrics for each instance, and find their average (only meaningful for multilabel classification).
57
  sample_weight (`list` of `float`): Sample weights Defaults to None.
58
- beta (`float`): Determines the weight of recall in the combined score.
59
-
60
  Returns:
61
- F1 (`float` (if average is not None) or `array` of `float`, shape =\ [n_unique_labels]): score of the positive class in binary classification or weighted average of the F1 scores of each class for the multiclass task.
62
  The F-beta score is the weighted harmonic mean of precision and recall, reaching its optimal value at 1 and its worst value at 0.
63
 
64
- Examples:
65
-
66
- >>> f_beta = evaluate.load("leslyarun/f_beta")
67
- >>> results = f_beta.compute(references=[0, 1], predictions=[0, 1])
68
- >>> print(results)
69
- {'f_beta_score': 1.0}
70
 
71
- For further examples, refer to https://scikit-learn.org/stable/modules/generated/sklearn.metrics.fbeta_score.html#sklearn.metrics.fbeta_score
 
 
 
 
72
  """
73
 
74
 
@@ -84,9 +83,9 @@ class F_Beta(evaluate.Metric):
84
  # This defines the format of each prediction and reference
85
  features=datasets.Features({
86
  'predictions': datasets.Value('int32'),
87
- 'references': datasets.Value('int32'),
88
- 'beta': datasets.Value('float32')
89
  }),
 
90
  # Homepage of the module for documentation
91
  homepage="https://huggingface.co/spaces/leslyarun/fbeta_score",
92
  # Additional links to the codebase or references
 
46
  Args:
47
  predictions (`list` of `int`): Predicted labels.
48
  references (`list` of `int`): Ground truth labels.
49
+ beta (`float`): Determines the weight of recall in the combined score.
50
  labels (`list` of `int`): The set of labels to include when `average` is not set to `'binary'`, and the order of the labels if `average` is `None`. Labels present in the data can be excluded, for example to calculate a multiclass average ignoring a majority negative class. Labels not present in the data will result in 0 components in a macro average. For multilabel targets, labels are column indices. By default, all labels in `predictions` and `references` are used in sorted order. Defaults to None.
51
  pos_label (`int`): The class to be considered the positive class, in the case where `average` is set to `binary`. Defaults to 1.
52
  average (`string`): This parameter is required for multiclass/multilabel targets. If set to `None`, the scores for each class are returned. Otherwise, this determines the type of averaging performed on the data. Defaults to `'binary'`.
 
56
  - 'weighted': Calculate metrics for each label, and find their average weighted by support (the number of true instances for each label). This alters `'macro'` to account for label imbalance. This option can result in an F-score that is not between precision and recall.
57
  - 'samples': Calculate metrics for each instance, and find their average (only meaningful for multilabel classification).
58
  sample_weight (`list` of `float`): Sample weights Defaults to None.
59
+
 
60
  Returns:
61
+ fbeta_score (`float` (if average is not None) or `array` of `float`, shape =\ [n_unique_labels]): of the positive class in binary classification or weighted average of the F1 scores of each class for the multiclass task.
62
  The F-beta score is the weighted harmonic mean of precision and recall, reaching its optimal value at 1 and its worst value at 0.
63
 
64
+ Examples:
 
 
 
 
 
65
 
66
+ Example 1-A simple binary example
67
+ >>> f_beta = evaluate.load("leslyarun/f_beta")
68
+ >>> results = f_beta.compute(references=[0, 1], predictions=[0, 1])
69
+ >>> print(results)
70
+ {'f_beta_score': 1.0}
71
  """
72
 
73
 
 
83
  # This defines the format of each prediction and reference
84
  features=datasets.Features({
85
  'predictions': datasets.Value('int32'),
86
+ 'references': datasets.Value('int32')
 
87
  }),
88
+ 'beta': datasets.Value('float32')
89
  # Homepage of the module for documentation
90
  homepage="https://huggingface.co/spaces/leslyarun/fbeta_score",
91
  # Additional links to the codebase or references