add docs for fairness

docs/api/metrics.rst

@@ -17,3 +17,4 @@ tasks, such as drug drug interaction (DDI) rate.
  metrics/pyhealth.metrics.binary
  metrics/pyhealth.metrics.calibration
  metrics/pyhealth.metrics.prediction_set
+ metrics/pyhealth.metrics.fairness

docs/api/metrics/pyhealth.metrics.fairness.rst

@@ -0,0 +1,14 @@
+pyhealth.metrics.fairness
+===================================
+
+.. currentmodule:: pyhealth.metrics.fairness
+
+.. autofunction:: fairness_metrics_fn
+
+.. currentmodule:: pyhealth.metrics.fairness_utils
+
+.. autofunction:: disparate_impact
+
+.. autofunction:: statistical_parity_difference
+
+.. autofunction:: sensitive_attributes_from_patient_ids

docs/log.rst

@@ -2,6 +2,12 @@ Development logs
 ======================
 We track the new development here:
 
+**Sep 01, 2023**
+
+.. code-block:: rst
+
+ 1. add Base fairness metrics and example `#216`.
+
 **July 22, 2023**
 
 .. code-block:: rst

pyhealth/metrics/fairness_utils/group.py

@@ -11,11 +11,13 @@ def disparate_impact(sensitive_attributes: np.ndarray, y_pred: np.ndarray, favor
  Computes the disparate impact between the the protected and unprotected group.
 
  disparate_impact = P(y_pred = favorable_outcome | P) / P(y_pred = favorable_outcome | U)
+ 
  Args:
  sensitive_attributes: Sensitive attributes of shape (n_samples,) where 1 is the protected group and 0 is the unprotected group.
  y_pred: Predicted target values of shape (n_samples,).
  favorable_outcome: Label value which is considered favorable (i.e. "positive").
  allow_zero_division: If True, use epsilon instead of 0 in the denominator if the denominator is 0. Otherwise, raise a ValueError.
+ 
  Returns:
  The disparate impact between the protected and unprotected group.
  """