make docs clear about FR corrected stat

hyppo/independence/friedman_rafsky.py

@@ -1,11 +1,19 @@
 import random
+from typing import NamedTuple
+
 from numba import jit
 import numpy as np
 
-from .base import IndependenceTest, IndependenceTestOutput
+from .base import IndependenceTest
 from ..tools import perm_test
 
 
+class FRTestOutput(NamedTuple):
+ stat: float
+ pvalue: float
+ uncor_stat: dict
+
+
 class FriedmanRafsky(IndependenceTest):
  r"""
  Friedman-Rafksy (FR) test statistic and p-value.
@@ -35,41 +43,19 @@ class are removed. The number of independent graphs is then summed to determine
 
  The p-value and null distribution for the corrected statistic are calculated via
  a permutation test using :meth:`hyppo.tools.perm_test`.
+
+ References
+ ----------
+ .. footbibliography::
  """
 
  def __init__(self, **kwargs):
 
  IndependenceTest.__init__(self, **kwargs)
 
- def _num_runs(self, labels, MST_connections):
- r"""
- Helper function to determine number of independent
- 'runs' from MST connections.
-
- Parameters
- ----------
- labels : ndarry of float
- Lables corresponding to respective classes of samples.
- MST_connections: list of int
- List containing pairs of points connected in final MST.
-
- Returns
- -------
- run_count : int
- Number of runs after severing all such edges with nodes of
- differing class labels.
- """
- run_count = 1
-
- for x in MST_connections:
- if labels[x[0]] != labels[x[1]]:
- run_count += 1
-
- return run_count
-
  def statistic(self, x, y):
  r"""
- Helper function that calculates the Friedman Rafksy test statistic. 
+ Helper function that calculates the Friedman Rafksy test statistic.
 
  Parameters
  ----------
@@ -83,13 +69,14 @@ def statistic(self, x, y):
  Returns
  -------
  stat : float
- The computed Friedman Rafsky statistic. A value between ``2`` and ``n``.
+ The computed (uncorrected) Friedman Rafsky statistic. A value between
+ ``2`` and ``n``.
  """
  x = np.transpose(x)
  labels = np.transpose(y)
 
  MST_connections = MST(x, labels)
- stat = self._num_runs(labels, MST_connections)
+ stat = _num_runs(labels, MST_connections)
 
  return stat
 
@@ -102,7 +89,7 @@ def test(
  random_state=None,
  ):
  r"""
- Calculates the Friedman Rafsky test statistic and p-value. 
+ Calculates the Friedman Rafsky test statistic and p-value.
 
  Parameters
  ----------
@@ -125,12 +112,13 @@ def test(
  Returns
  -------
  stat : float
- The computed Friedman Rafsky statistic.
+ The computed (corrected) Friedman Rafsky statistic.
  pvalue : float
  The computed Friedman Rafsky p-value.
+ uncor_stat : float
+ The computed (uncorrected) Friedman Rafsky statistic.
  """
-
- stat, pvalue, null_dist = perm_test(
+ uncor_stat, pvalue, null_dist = perm_test(
  self.statistic,
  x,
  y,
@@ -139,10 +127,38 @@ def test(
  is_distsim=False,
  random_state=random_state,
  )
- stat = (stat - np.mean(null_dist)) / np.std(null_dist)
+ self.uncor_stat = uncor_stat
+ stat = (uncor_stat - np.mean(null_dist)) / np.std(null_dist)
  self.stat = stat
 
- return IndependenceTestOutput(stat, pvalue)
+ return FRTestOutput(stat, pvalue, uncor_stat)
+
+
+def _num_runs(labels, MST_connections):
+ r"""
+ Helper function to determine number of independent
+ 'runs' from MST connections.
+
+ Parameters
+ ----------
+ labels : ndarry of float
+ Lables corresponding to respective classes of samples.
+ MST_connections: list of int
+ List containing pairs of points connected in final MST.
+
+ Returns
+ -------
+ run_count : int
+ Number of runs after severing all such edges with nodes of
+ differing class labels.
+ """
+ run_count = 1
+
+ for x in MST_connections:
+ if labels[x[0]] != labels[x[1]]:
+ run_count += 1
+
+ return run_count
 
 
 @jit(nopython=True, cache=True) # pragma: no cover

hyppo/independence/mgc.py

@@ -9,7 +9,7 @@
 from .base import IndependenceTest
 
 
-class MGCestOutput(NamedTuple):
+class MGCTestOutput(NamedTuple):
  stat: float
  pvalue: float
  mgc_dict: dict
@@ -259,4 +259,4 @@ def test(self, x, y, reps=1000, workers=1, random_state=None):
  )
  self.mgc_dict = mgc_dict
 
- return MGCestOutput(stat, pvalue, mgc_dict)
+ return MGCTestOutput(stat, pvalue, mgc_dict)

hyppo/independence/tests/test_friedman_rafsky.py

@@ -17,7 +17,7 @@ def test_linear_oned(self, n, num_runs, obs_stat, obs_pvalue):
  num_rows, num_cols = x.shape
  y = np.random.choice([0, 1], num_rows, p=[0.5, 0.5])
  y = np.transpose(y)
- stat1, pvalue1 = FriedmanRafsky().test(x, y)
+ stat1, pvalue1, _ = FriedmanRafsky().test(x, y)
  stat2 = FriedmanRafsky().statistic(x, y)
 
  assert_almost_equal(stat1, obs_stat, decimal=2)
@@ -30,8 +30,8 @@ def test_rep(self, n):
  num_rows, num_cols = x.shape
  y = np.random.choice([0, 1], num_rows, p=[0.5, 0.5])
  y = np.transpose(y)
- stat1, pvalue1 = FriedmanRafsky().test(x, y, random_state=2)
- stat2, pvalue2 = FriedmanRafsky().test(x, y, random_state=2)
+ stat1, pvalue1, _ = FriedmanRafsky().test(x, y, random_state=2)
+ stat2, pvalue2, _ = FriedmanRafsky().test(x, y, random_state=2)
 
  assert stat1 == stat2
  assert pvalue1 == pvalue2