def test_select_merge_components_random_raisesError(self):
''' Verify that when comp 0 is excluded with K=3
we cannot provide comp 0 as kA, [error is raised]
AND
in free choice, we only choose kA=1, kB=2
'''
MT = MergeTracker(3)
MSelector = MergePairSelector()
MT.excludeList = set([0])
MT._synchronize_and_verify()
for trial in range(10):
kA, kB = MSelector.select_merge_components(None,
None,
MT,
kA=1,
mergename='random')
assert kA == 1
assert kB == 2
for trial in range(10):
kA, kB = MSelector.select_merge_components(None,
None,
MT,
kA=2,
mergename='random')
assert kA == 1
assert kB == 2
with self.assertRaises(AssertionError):
kA, kB = MSelector.select_merge_components(None,
None,
MT,
mergename='random',
kA=0)
def test_select_merge_components_random_raisesErrorAllButOneExcluded(self):
''' Verify that when comps 0,1 are excluded with K=3
we cannot provide comp 2 as kA, [error is raised]
'''
MT = MergeTracker(3)
MSelector = MergePairSelector()
MT.excludeList = set([1, 0])
MT._synchronize_and_verify()
with self.assertRaises(AssertionError):
kA, kB = MSelector.select_merge_components(None,
None,
MT,
mergename='random',
kA=2)
def test_reindexAfterMerge(self):
MSelector = MergePairSelector()
MSelector.MScores[0] = 5
MSelector.MScores[3] = 5
MSelector.MScores[4] = 5
MSelector.PairMScores[(0, 1)] = 5
MSelector.PairMScores[(3, 4)] = 5
MSelector.PairMScores[(5, 6)] = 5
MSelector.reindexAfterMerge(2, 3)
assert MSelector.MScores[0] == 5
assert MSelector.MScores[3] == 5
assert 2 not in MSelector.MScores
assert len(MSelector.PairMScores.keys()) == 2
assert (0, 1) in MSelector.PairMScores
assert (4, 5) in MSelector.PairMScores
assert (2, 3) not in MSelector.PairMScores
def test_select_merge_components_random(self):
''' Verify that under random choices, we select among 3 components
equally often
'''
MT = MergeTracker(3)
MSelector = MergePairSelector()
counts = np.zeros(3)
for trial in range(1000):
kA, kB = MSelector.select_merge_components(None,
None,
MT,
mergename='random')
counts[kA] += 1
counts[kB] += 1
counts /= np.sum(counts)
minFrac = 0.25
maxFrac = 0.4
# Uniform at random means fraction of choice should be ~1/3 for each
assert np.all(counts > minFrac)
assert np.all(counts < maxFrac)
