implement specificity

README.md

@@ -1,11 +1,9 @@
 ---
 title: Specificity
-datasets:
--  
 tags:
 - evaluate
 - metric
-description: "TODO: add a description here"
+description: "Specificity is the fraction of the negatives examples that were correctly labeled by the model as negatives. It can be computed with the equation: Specificity = TN / (TN + FP) Where TN is the true negatives and FP is the false positives."
 sdk: gradio
 sdk_version: 3.19.1
 app_file: app.py
@@ -14,37 +12,96 @@ pinned: false
 
 # Metric Card for Specificity
 
-***Module Card Instructions:*** *Fill out the following subsections. Feel free to take a look at existing metric cards if you'd like examples.*
-
 ## Metric Description
-*Give a brief overview of this metric, including what task(s) it is usually used for, if any.*
+
+Specificity is the fraction of the negatives examples that were correctly labeled by the model as negatives. It can be computed with the equation:
+$$\text{Specificity} = \frac{TN}{TN + FP}$$
+Where $TN$ is the true negatives and $FP$ is the false positives.
 
 ## How to Use
-*Give general statement of how to use the metric*
 
-*Provide simplest possible example for using the metric*
+At minimum, this metric takes as input two lists, each containing ints: predictions and references.
+
+```python
+>>> specificity_metric = evaluate.load('nevikw39/specificity')
+>>> results = specificity_metric.compute(references=[0, 1], predictions=[0, 1])
+>>> print(results)
+["{'specificity': 1.0}"]
+```
 
 ### Inputs
-*List all input arguments in the format below*
-- **input_field** *(type): Definition of input, with explanation if necessary. State any default value(s).*
 
-### Output Values
+- **predictions** (`list` of `int`): The predicted labels.
+- **references** (`list` of `int`): The ground truth labels.
+- **labels** (`list` of `int`): The set of labels to include when `average` is not set to `binary`, and their order when average is `None`. Labels present in the data can be excluded in this input, for example to calculate a multiclass average ignoring a majority negative class, while 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 y_true and y_pred are used in sorted order. Defaults to None.
+- **neg_label** (`int`): The class label to use as the 'negative class' when calculating the specificity. Defaults to `0`.
+- **average** (`string`): This parameter is required for multiclass/multilabel targets. If None, the scores for each class are returned. Otherwise, this determines the type of averaging performed on the data. Defaults to `'binary'`.
+    - `'binary'`: Only report results for the class specified by `neg_label`. This is applicable only if the target labels and predictions are binary.
+    - `'micro'`: Calculate metrics globally by counting the total true negatives, false positives, and false negatives.
+    - `'macro'`: Calculate metrics for each label, and find their unweighted mean. This does not take label imbalance into account.
+    - `'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. Note that it can result in an F-score that is not between sensitivity and specificity.
+    - `'samples'`: Calculate metrics for each instance, and find their average (only meaningful for multilabel classification).
+- **sample_weight** (`list` of `float`): Sample weights Defaults to `None`.
+- **zero_division** (): Sets the value to return when there is a zero division. Defaults to .
+    - `'warn'`: If there is a zero division, the return value is `0`, but warnings are also raised.
+    - `0`: If there is a zero division, the return value is `0`.
+    - `1`: If there is a zero division, the return value is `1`.
 
-*Explain what this metric outputs and provide an example of what the metric output looks like. Modules should return a dictionary with one or multiple key-value pairs, e.g. {"bleu" : 6.02}*
+### Output Values
 
-*State the range of possible values that the metric's output can take, as well as what in that range is considered good. For example: "This metric can take on any value between 0 and 100, inclusive. Higher scores are better."*
+- **specificity** (`float`, or `array` of `float`): Either the general specificity score, or the specificity scores for individual classes, depending on the values input to `labels` and `average`. Minimum possible value is 0. Maximum possible value is 1. A higher specificity means that more of the positive examples have been labeled correctly. Therefore, a higher specificity is generally considered better.
 
 #### Values from Popular Papers
-*Give examples, preferrably with links to leaderboards or publications, to papers that have reported this metric, along with the values they have reported.*
 
 ### Examples
-*Give code examples of the metric being used. Try to include examples that clear up any potential ambiguity left from the metric description above. If possible, provide a range of examples that show both typical and atypical results, as well as examples where a variety of input parameters are passed.*
+
+Example 1-A simple example with some errors
+```python
+    >>> specificity_metric = evaluate.load('nevikw39/specificity')
+    >>> results = specificity_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1])
+    >>> print(results)
+    {'specificity': 0.5}
+```
+Example 2-The same example as Example 1, but with `neg_label=1` instead of the default `neg_label=0`.
+```python
+    >>> specificity_metric = evaluate.load('nevikw39/specificity')
+    >>> results = specificity_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1], neg_label=1)
+    >>> print(results)
+    {'specificity': 0.6666666666666666}
+```
+Example 3-The same example as Example 1, but with `sample_weight` included.
+```python
+    >>> specificity_metric = evaluate.load('nevikw39/specificity')
+    >>> sample_weight = [0.9, 0.2, 0.9, 0.3, 0.8]
+    >>> results = specificity_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1], sample_weight=sample_weight)
+    >>> print(results)
+    {'specificity': 0.8181818181818181}
+```
+Example 4-A multiclass example, using different averages.
+```python
+    >>> specificity_metric = evaluate.load('nevikw39/specificity')
+    >>> predictions = [0, 2, 1, 0, 0, 1]
+    >>> references = [0, 1, 2, 0, 1, 2]
+    >>> results = specificity_metric.compute(predictions=predictions, references=references, average='macro')
+    >>> print(results)
+    {'specificity': 0.3333333333333333}
+    >>> results = specificity_metric.compute(predictions=predictions, references=references, average='micro')
+    >>> print(results)
+    {'specificity': 0.3333333333333333}
+    >>> results = specificity_metric.compute(predictions=predictions, references=references, average='weighted')
+    >>> print(results)
+    {'specificity': 0.3333333333333333}
+    >>> results = specificity_metric.compute(predictions=predictions, references=references, average=None)
+    >>> print(results)
+    {'specificity': array([1., 0., 0.])}
+```
 
 ## Limitations and Bias
-*Note any known limitations or biases that the metric has, with links and references if possible.*
 
 ## Citation
-*Cite the source where this metric was introduced.*
+
+```
+@article{scikit-learn, title={Scikit-learn: Machine Learning in {P}ython}, author={Pedregosa, F. and Varoquaux, G. and Gramfort, A. and Michel, V. and Thirion, B. and Grisel, O. and Blondel, M. and Prettenhofer, P. and Weiss, R. and Dubourg, V. and Vanderplas, J. and Passos, A. and Cournapeau, D. and Brucher, M. and Perrot, M. and Duchesnay, E.}, journal={Journal of Machine Learning Research}, volume={12}, pages={2825--2830}, year={2011}}
+```
 
 ## Further References
-*Add any useful further references.*

specificity.py

@@ -11,85 +11,124 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-"""TODO: Add a description here."""
+"""Specificity metric."""
 
 import evaluate
 import datasets
 
+from sklearn.metrics import recall_score
+
 
-# TODO: Add BibTeX citation
 _CITATION = """\
-@InProceedings{huggingface:module,
-title = {A great new module},
-authors={huggingface, Inc.},
-year={2020}
-}
+@article{scikit-learn, title={Scikit-learn: Machine Learning in {P}ython}, author={Pedregosa, F. and Varoquaux, G. and Gramfort, A. and Michel, V. and Thirion, B. and Grisel, O. and Blondel, M. and Prettenhofer, P. and Weiss, R. and Dubourg, V. and Vanderplas, J. and Passos, A. and Cournapeau, D. and Brucher, M. and Perrot, M. and Duchesnay, E.}, journal={Journal of Machine Learning Research}, volume={12}, pages={2825--2830}, year={2011}}
 """
 
-# TODO: Add description of the module here
+
 _DESCRIPTION = """\
-This new module is designed to solve this great ML task and is crafted with a lot of care.
+Specificity is the fraction of the negatives examples that were correctly labeled by the model as negatives. It can be computed with the equation:
+Specificity = TN / (TN + FP)
+Where TN is the true negatives and FP is the false positives.
 """
 
 
-# TODO: Add description of the arguments of the module here
 _KWARGS_DESCRIPTION = """
-Calculates how good are predictions given some references, using certain scores
 Args:
-    predictions: list of predictions to score. Each predictions
-        should be a string with tokens separated by spaces.
-    references: list of reference for each prediction. Each
-        reference should be a string with tokens separated by spaces.
+- **predictions** (`list` of `int`): The predicted labels.
+- **references** (`list` of `int`): The ground truth labels.
+- **labels** (`list` of `int`): The set of labels to include when `average` is not set to `binary`, and their order when average is `None`. Labels present in the data can be excluded in this input, for example to calculate a multiclass average ignoring a majority negative class, while 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 y_true and y_pred are used in sorted order. Defaults to None.
+- **neg_label** (`int`): The class label to use as the 'negative class' when calculating the specificity. Defaults to `0`.
+- **average** (`string`): This parameter is required for multiclass/multilabel targets. If None, the scores for each class are returned. Otherwise, this determines the type of averaging performed on the data. Defaults to `'binary'`.
+    - `'binary'`: Only report results for the class specified by `neg_label`. This is applicable only if the target labels and predictions are binary.
+    - `'micro'`: Calculate metrics globally by counting the total true negatives, false positives, and false negatives.
+    - `'macro'`: Calculate metrics for each label, and find their unweighted mean. This does not take label imbalance into account.
+    - `'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. Note that it can result in an F-score that is not between sensitivity and specificity.
+    - `'samples'`: Calculate metrics for each instance, and find their average (only meaningful for multilabel classification).
+- **sample_weight** (`list` of `float`): Sample weights Defaults to `None`.
+- **zero_division** (): Sets the value to return when there is a zero division. Defaults to .
+    - `'warn'`: If there is a zero division, the return value is `0`, but warnings are also raised.
+    - `0`: If there is a zero division, the return value is `0`.
+    - `1`: If there is a zero division, the return value is `1`.
 Returns:
-    accuracy: description of the first score,
-    another_score: description of the second score,
+- **specificity** (`float`, or `array` of `float`): Either the general specificity score, or the specificity scores for individual classes, depending on the values input to `labels` and `average`. Minimum possible value is 0. Maximum possible value is 1. A higher specificity means that more of the positive examples have been labeled correctly. Therefore, a higher specificity is generally considered better.
 Examples:
-    Examples should be written in doctest format, and should illustrate how
-    to use the function.
-
-    >>> my_new_module = evaluate.load("my_new_module")
-    >>> results = my_new_module.compute(references=[0, 1], predictions=[0, 1])
-    >>> print(results)
-    {'accuracy': 1.0}
+    Example 1-A simple example with some errors
+        >>> specificity_metric = evaluate.load('nevikw39/specificity')
+        >>> results = specificity_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1])
+        >>> print(results)
+        {'specificity': 0.5}
+    Example 2-The same example as Example 1, but with `neg_label=1` instead of the default `neg_label=0`.
+        >>> specificity_metric = evaluate.load('nevikw39/specificity')
+        >>> results = specificity_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1], neg_label=1)
+        >>> print(results)
+        {'specificity': 0.6666666666666666}
+    Example 3-The same example as Example 1, but with `sample_weight` included.
+        >>> specificity_metric = evaluate.load('nevikw39/specificity')
+        >>> sample_weight = [0.9, 0.2, 0.9, 0.3, 0.8]
+        >>> results = specificity_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1], sample_weight=sample_weight)
+        >>> print(results)
+        {'specificity': 0.8181818181818181}
+    Example 4-A multiclass example, using different averages.
+        >>> specificity_metric = evaluate.load('nevikw39/specificity')
+        >>> predictions = [0, 2, 1, 0, 0, 1]
+        >>> references = [0, 1, 2, 0, 1, 2]
+        >>> results = specificity_metric.compute(predictions=predictions, references=references, average='macro')
+        >>> print(results)
+        {'specificity': 0.3333333333333333}
+        >>> results = specificity_metric.compute(predictions=predictions, references=references, average='micro')
+        >>> print(results)
+        {'specificity': 0.3333333333333333}
+        >>> results = specificity_metric.compute(predictions=predictions, references=references, average='weighted')
+        >>> print(results)
+        {'specificity': 0.3333333333333333}
+        >>> results = specificity_metric.compute(predictions=predictions, references=references, average=None)
+        >>> print(results)
+        {'specificity': array([1., 0., 0.])}
 """
 
-# TODO: Define external resources urls if needed
-BAD_WORDS_URL = "http://url/to/external/resource/bad_words.txt"
-
 
 @evaluate.utils.file_utils.add_start_docstrings(_DESCRIPTION, _KWARGS_DESCRIPTION)
 class Specificity(evaluate.Metric):
-    """TODO: Short description of my evaluation module."""
+    """Specificity metric."""
 
     def _info(self):
-        # TODO: Specifies the evaluate.EvaluationModuleInfo object
         return evaluate.MetricInfo(
-            # This is the description that will appear on the modules page.
             module_type="metric",
             description=_DESCRIPTION,
             citation=_CITATION,
             inputs_description=_KWARGS_DESCRIPTION,
-            # This defines the format of each prediction and reference
-            features=datasets.Features({
-                'predictions': datasets.Value('int64'),
-                'references': datasets.Value('int64'),
-            }),
-            # Homepage of the module for documentation
-            homepage="http://module.homepage",
-            # Additional links to the codebase or references
-            codebase_urls=["http://github.com/path/to/codebase/of/new_module"],
-            reference_urls=["http://path.to.reference.url/new_module"]
+            features=datasets.Features(
+                {
+                    "predictions": datasets.Sequence(datasets.Value("int32")),
+                    "references": datasets.Sequence(datasets.Value("int32")),
+                }
+                if self.config_name == "multilabel"
+                else {
+                    "predictions": datasets.Value("int32"),
+                    "references": datasets.Value("int32"),
+                }
+            ),
+            reference_urls=[
+                "https://scikit-learn.org/stable/modules/generated/sklearn.metrics.recall_score.html"
+            ],
         )
 
-    def _download_and_prepare(self, dl_manager):
-        """Optional: download external resources useful to compute the scores"""
-        # TODO: Download external resources if needed
-        pass
-
-    def _compute(self, predictions, references):
-        """Returns the scores"""
-        # TODO: Compute the different scores of the module
-        accuracy = sum(i == j for i, j in zip(predictions, references)) / len(predictions)
-        return {
-            "accuracy": accuracy,
-        }
+    def _compute(
+        self,
+        predictions: datasets.Sequence(datasets.Value("int32")) | datasets.Value("int32"),
+        references: datasets.Sequence(datasets.Value("int32")) | datasets.Value("int32"),
+        labels=None,
+        neg_label=0,
+        average="binary",
+        sample_weight=None,
+        zero_division="warn",
+    ):
+        score = recall_score(
+            references,
+            predictions,
+            labels=labels,
+            pos_label=neg_label,
+            average=average,
+            sample_weight=sample_weight,
+            zero_division=zero_division,
+        )
+        return {"specificity": float(score) if score.size == 1 else score}

tests.py

@@ -1,17 +1,7 @@
 test_cases = [
     {
-        "predictions": [0, 0],
-        "references": [1, 1],
-        "result": {"metric_score": 0}
+        "predictions": [0, 0, 1, 1, 1],
+        "references": [0, 1, 0, 1, 1],
+        "result": {"specificity": 0.5}
     },
-    {
-        "predictions": [1, 1],
-        "references": [1, 1],
-        "result": {"metric_score": 1}
-    },
-    {
-        "predictions": [1, 0],
-        "references": [1, 1],
-        "result": {"metric_score": 0.5}
-    }
 ]