def test_nltk(self):
'''
Runs WD tests from https://github.com/nltk/nltk/blob/master/nltk/test/segmentation.doctest
'''
s1 = "000100000010"
s2 = "000010000100"
s3 = "100000010000"
# Originally 0.0
self.assertAlmostEqual(
window_diff(
s1,
s1,
window_size=3,
boundary_format=BoundaryFormat.nltk),
Decimal('0'))
# Originally 0.3
self.assertAlmostEqual(
window_diff(
s2,
s1,
window_size=3,
boundary_format=BoundaryFormat.nltk),
Decimal('0.3'))
# Originally 0.7
self.assertAlmostEqual(
window_diff(
s3,
s2,
window_size=3,
boundary_format=BoundaryFormat.nltk),
Decimal('0.8'))
def test_nltk(self):
'''
Runs WD tests from https://github.com/nltk/nltk/blob/master/nltk/test/segmentation.doctest
'''
s1 = "000100000010"
s2 = "000010000100"
s3 = "100000010000"
# Originally 0.0
self.assertAlmostEqual(
window_diff(s1,
s1,
window_size=3,
boundary_format=BoundaryFormat.nltk), Decimal('0'))
# Originally 0.3
self.assertAlmostEqual(
window_diff(s2,
s1,
window_size=3,
boundary_format=BoundaryFormat.nltk), Decimal('0.3'))
# Originally 0.7
self.assertAlmostEqual(
window_diff(s3,
s2,
window_size=3,
boundary_format=BoundaryFormat.nltk), Decimal('0.8'))
def test_all_and_no_boundaries(self):
'''
Test whether all segments versus no segments produces 1.0.
'''
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
b = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs), Decimal('1'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs), Decimal('1'))
def test_all_and_no_boundaries(self):
'''
Test whether all segments versus no segments produces 1.0.
'''
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
b = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs), Decimal('1'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs), Decimal('1'))
def test_no_boundaries(self):
'''
Test whether no segments versus some segments produce 1.0.
'''
a = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.3636363636363636363636363636'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs), Decimal('1'))
def test_extra_boundary(self):
'''
Test extra boundary.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('2.0') / Decimal('11.0'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('2.0') / Decimal('11.0'))
def test_no_boundaries(self):
'''
Test whether no segments versus some segments produce 1.0.
'''
a = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.3636363636363636363636363636'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('1'))
def test_translated_boundary(self):
'''
Test mis-alignment.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs), # k = 2
Decimal('2.0') / Decimal('11.0'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('2.0') / Decimal('11.0'))
def test_extra_boundary(self):
'''
Test extra boundary.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('2.0') / Decimal('11.0'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('2.0') / Decimal('11.0'))
def test_full_miss_and_misaligned(self):
'''
Test whether a full miss and a translated boundary out of 4 produces
0.25.
'''
a = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.2727272727272727272727272727'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('0.2727272727272727272727272727'))
def test_translated_boundary(self):
'''
Test mis-alignment.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(
window_diff(a, b, **self.kwargs), # k = 2
Decimal('2.0') / Decimal('11.0'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('2.0') / Decimal('11.0'))
def test_all_boundaries(self):
'''
Test whether all segments versus some segments produces 0.833
erroneous windows.
'''
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('1.0'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('1.0'))
def test_all_boundaries(self):
'''
Test whether all segments versus some segments produces 0.833
erroneous windows.
'''
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('1.0'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('1.0'))
def test_full_miss_and_misaligned(self):
'''
Test whether a full miss and a translated boundary out of 4 produces
0.25.
'''
a = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.2727272727272727272727272727'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('0.2727272727272727272727272727'))
def test_one_minus(self):
'''
Test whether no segments versus some segments produce 1.0.
'''
a = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
metric_kwargs = dict(self.kwargs)
metric_kwargs['one_minus'] = True
self.assertAlmostEqual(window_diff(a, b, **metric_kwargs),
Decimal('0.6363636363636363636363636364'))
self.assertAlmostEqual(window_diff(b, a, **metric_kwargs),
Decimal('0'))
def test_one_minus(self):
'''
Test whether no segments versus some segments produce 1.0.
'''
a = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
metric_kwargs = dict(self.kwargs)
metric_kwargs['one_minus'] = True
self.assertAlmostEqual(window_diff(a, b, **metric_kwargs),
Decimal('0.6363636363636363636363636364'))
self.assertAlmostEqual(window_diff(b, a, **metric_kwargs),
Decimal('0'))
def test_complete_agreement(self):
'''
Calculate mean permuted pairwise WindowDiff on a theoretical dataset
containing complete agreement.
'''
self.assertAlmostEquals(
summarize(
window_diff(COMPLETE_AGREEMENT,
lamprier_et_al_2007_fix=False)),
(0.0, 0.0, 0.0, 0.0, 48))
self.assertAlmostEquals(
summarize(
window_diff(COMPLETE_AGREEMENT, lamprier_et_al_2007_fix=True)),
(0.0, 0.0, 0.0, 0.0, 48))
def test_long_format(self):
hypothesis = (2, 31, 4, 1, 1, 3, 11, 5, 21, 4, 2, 1, 17, 26, 16, 1, 17,
4, 3, 7, 7, 6, 12, 1, 6, 25, 2, 4, 3, 16, 8)
reference = (2, 36, 1, 3, 10, 1, 5, 21, 4, 3, 59, 8, 10, 4, 3, 7, 13,
12, 7, 27, 4, 3, 24)
self.assertAlmostEqual(window_diff(hypothesis, reference),
Decimal('0.2681992337164750957854406130'))
def test_long_format(self):
hypothesis = (
2,
31,
4,
1,
1,
3,
11,
5,
21,
4,
2,
1,
17,
26,
16,
1,
17,
4,
3,
7,
7,
6,
12,
1,
6,
25,
2,
4,
3,
16,
8)
reference = (
2,
36,
1,
3,
10,
1,
5,
21,
4,
3,
59,
8,
10,
4,
3,
7,
13,
12,
7,
27,
4,
3,
24)
self.assertAlmostEqual(
window_diff(hypothesis, reference),
Decimal('0.2681992337164750957854406130'))
def test_identical(self):
'''
Test whether identical segmentations produce 0.0.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertEqual(window_diff(a, b, **self.kwargs), Decimal('0.0'))
def test_fn_vs_fp(self):
'''
Test the difference between FP and FN.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
a = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
b = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
def test_fn_vs_fp(self):
'''
Test the difference between FP and FN.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
a = [1, 1, 1, 1, 1, 2, 3, 3, 4, 4, 4, 4, 4]
b = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertAlmostEqual(window_diff(a, b, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
self.assertAlmostEqual(window_diff(b, a, **self.kwargs),
Decimal('0.1818181818181818181818181818'))
def test_identical(self):
'''
Test whether identical segmentations produce 0.0.
'''
a = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
b = [1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3]
self.assertEqual(window_diff(a, b, **self.kwargs), Decimal('0.0'))
def test_kazantseva2012_g5(self):
'''
Calculate permuted pairwise WindowDiff on Group 5 from the dataset
collected in [KazantsevaSzpakowicz2012]_.
'''
self.assertAlmostEquals(
summarize(
window_diff(KAZANTSEVA2012_G5, lamprier_et_al_2007_fix=False)),
(Decimal('0.42514977232721135'), Decimal('0.14960495739111837'),
Decimal('0.02238164327599834'), Decimal('0.02159361560546617'),
48))
self.assertAlmostEquals(
summarize(
window_diff(KAZANTSEVA2012_G5, lamprier_et_al_2007_fix=True)),
(Decimal('0.3981624488837195'), Decimal('0.1499358110615409'),
Decimal('0.022480747438682093'), Decimal('0.021641370219386377'),
48))
def test_thesis_all_and_none(self):
'''
Test paper example A vs B
'''
reference = [11]
hypothesis = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
# Test normal
actual = window_diff(hypothesis, reference, **self.kwargs)
self.assertAlmostEqual(1.0, float(actual))
def test_large_disagreement(self):
'''
Calculate mean permuted pairwise WindowDiff on a theoretical dataset
containing large disagreement.
'''
val = summarize(
window_diff(LARGE_DISAGREEMENT, lamprier_et_al_2007_fix=False))
self.assertAlmostEquals(
val, (Decimal('1'), Decimal('0'), Decimal('0'), Decimal('0'), 8))
def test_boundary_format_nltk(self):
'''
Test the nltk boundary format.
'''
value = window_diff('0100100000',
'0101000000',
window_size=2,
boundary_format=BoundaryFormat.nltk)
self.assertAlmostEqual(Decimal('0.2222222'), value)
def test_scaiano_paper_e(self):
'''
Test paper example A vs E
'''
reference = [6, 6]
hypothesis = [5, 1, 1, 5]
# Test normal
actual = window_diff(hypothesis, reference, **self.kwargs)
self.assertAlmostEquals(5.0 / 9.0, float(actual))
def test_pair_g5(self):
'''
Test a comparison that is troublesome when using lamprier_et_al_2007_fix
'''
wd = window_diff(KAZANTSEVA2012_G5['ch4']['an2'],
KAZANTSEVA2012_G5['ch4']['an1'],
lamprier_et_al_2007_fix=True)
self.assertAlmostEqual(0.76363636, float(wd))
def test_thesis_all_and_none(self):
'''
Test paper example A vs B
'''
reference = [11]
hypothesis = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]
# Test normal
actual = window_diff(hypothesis, reference, **self.kwargs)
self.assertAlmostEqual(1.0, float(actual))
def test_scaiano_paper_e(self):
'''
Test paper example A vs E
'''
reference = [6, 6]
hypothesis = [5, 1, 1, 5]
# Test normal
actual = window_diff(hypothesis, reference, **self.kwargs)
self.assertAlmostEquals(5.0 / 9.0, float(actual))
def test_pair_g5(self):
'''
Test a comparison that is troublesome when using lamprier_et_al_2007_fix
'''
wd = window_diff(KAZANTSEVA2012_G5['ch4']['an2'],
KAZANTSEVA2012_G5['ch4']['an1'],
lamprier_et_al_2007_fix=True)
self.assertAlmostEqual(0.76363636, float(wd))
def test_kazantseva2012_g2(self):
'''
Calculate mean permuted pairwise WindowDiff on Group 2 from the dataset
collected in [KazantsevaSzpakowicz2012]_.
'''
self.assertAlmostEquals(
summarize(
window_diff(KAZANTSEVA2012_G2, lamprier_et_al_2007_fix=False)),
(Decimal('0.3257163091933661553914718469'),
Decimal('0.1586420969856167116081811670'),
Decimal('0.02516731493599381893573908435'),
Decimal('0.01448197584815743151147537110'), 120))
self.assertAlmostEquals(
summarize(
window_diff(KAZANTSEVA2012_G2, lamprier_et_al_2007_fix=True)),
(Decimal('0.2745037663246318112728760428'),
Decimal('0.1093940158628282748721971010'),
Decimal('0.01196705070659672423205913742'),
Decimal('0.009986261690691502298962284136'), 120))
def test_complete_agreement(self):
'''
Calculate mean permuted pairwise WindowDiff on a theoretical dataset
containing complete agreement.
'''
self.assertAlmostEquals(summarize(window_diff(COMPLETE_AGREEMENT,
lamprier_et_al_2007_fix=False)),
(0.0,
0.0,
0.0,
0.0,
48))
self.assertAlmostEquals(summarize(window_diff(COMPLETE_AGREEMENT,
lamprier_et_al_2007_fix=True)),
(0.0,
0.0,
0.0,
0.0,
48))
def test_kazantseva2012_g2(self):
'''
Calculate mean permuted pairwise WindowDiff on Group 2 from the dataset
collected in [KazantsevaSzpakowicz2012]_.
'''
self.assertAlmostEquals(summarize(window_diff(KAZANTSEVA2012_G2,
lamprier_et_al_2007_fix=False)),
(Decimal('0.3257163091933661553914718469'),
Decimal('0.1586420969856167116081811670'),
Decimal('0.02516731493599381893573908435'),
Decimal('0.01448197584815743151147537110'),
120))
self.assertAlmostEquals(summarize(window_diff(KAZANTSEVA2012_G2,
lamprier_et_al_2007_fix=True)),
(Decimal('0.2745037663246318112728760428'),
Decimal('0.1093940158628282748721971010'),
Decimal('0.01196705070659672423205913742'),
Decimal('0.009986261690691502298962284136'),
120))
def test_parts(self):
'''
Test parts.
'''
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
metric_kwargs = dict(self.kwargs)
metric_kwargs['return_parts'] = True
self.assertEqual(window_diff(a, b, **metric_kwargs), (11, 11))
def test_kazantseva2012_g5(self):
'''
Calculate permuted pairwise WindowDiff on Group 5 from the dataset
collected in [KazantsevaSzpakowicz2012]_.
'''
self.assertAlmostEquals(summarize(window_diff(KAZANTSEVA2012_G5,
lamprier_et_al_2007_fix=False)),
(Decimal('0.42514977232721135'),
Decimal('0.14960495739111837'),
Decimal('0.02238164327599834'),
Decimal('0.02159361560546617'),
48))
self.assertAlmostEquals(summarize(window_diff(KAZANTSEVA2012_G5,
lamprier_et_al_2007_fix=True)),
(Decimal('0.3981624488837195'),
Decimal('0.1499358110615409'),
Decimal('0.022480747438682093'),
Decimal('0.021641370219386377'),
48))
def test_boundary_format_nltk(self):
'''
Test the nltk boundary format.
'''
value = window_diff(
'0100100000',
'0101000000',
window_size=2,
boundary_format=BoundaryFormat.nltk)
self.assertAlmostEqual(Decimal('0.2222222'), value)
def test_parts(self):
'''
Test parts.
'''
a = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]
b = [1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3]
metric_kwargs = dict(self.kwargs)
metric_kwargs['return_parts'] = True
self.assertEqual(window_diff(a, b, **metric_kwargs),
(11, 11))
def test_wd_datasets(self):
'''
Test WD upon two datasets.
'''
hypothesis = HYPOTHESIS_STARGAZER
reference = HEARST_1997_STARGAZER
value = window_diff(hypothesis, reference)
# Precision
self.assertAlmostEquals(float(value['stargazer,h1,1']), 0.31578947)
self.assertAlmostEquals(float(value['stargazer,h2,1']), 0.47368421)
self.assertAlmostEquals(float(value['stargazer,h1,2']), 0.42105263)
self.assertAlmostEquals(float(value['stargazer,h2,2']), 0.47368421)
def test_large_disagreement(self):
'''
Calculate mean permuted pairwise WindowDiff on a theoretical dataset
containing large disagreement.
'''
val = summarize(window_diff(LARGE_DISAGREEMENT,
lamprier_et_al_2007_fix=False))
self.assertAlmostEquals(val,
(Decimal('1'),
Decimal('0'),
Decimal('0'),
Decimal('0'),
8))
def test_wd_datasets(self):
'''
Test WD upon two datasets.
'''
hypothesis = HYPOTHESIS_STARGAZER
reference = HEARST_1997_STARGAZER
value = window_diff(hypothesis, reference)
# Precision
self.assertAlmostEquals(float(value['stargazer,h1,1']), 0.31578947)
self.assertAlmostEquals(float(value['stargazer,h2,1']), 0.47368421)
self.assertAlmostEquals(float(value['stargazer,h1,2']), 0.42105263)
self.assertAlmostEquals(float(value['stargazer,h2,2']), 0.47368421)
def test_window_size_specified(self):
'''
Test when window size is specified.
'''
value = window_diff([2, 3, 6], [2, 2, 7], window_size=2)
self.assertAlmostEqual(Decimal('0.2222222'), value)
def test_window_size_specified(self):
'''
Test when window size is specified.
'''
value = window_diff([2, 3, 6], [2, 2, 7], window_size=2)
self.assertAlmostEqual(Decimal('0.2222222'), value)
def test_one_minus_var(self):
'''
Test one minus.
'''
value = window_diff([2, 3, 6], [2, 2, 7], one_minus=True)
self.assertAlmostEqual(Decimal('0.77777777'), value)
def test_one_minus_var(self):
'''
Test one minus.
'''
value = window_diff([2, 3, 6], [2, 2, 7], one_minus=True)
self.assertAlmostEqual(Decimal('0.77777777'), value)
