-
Notifications
You must be signed in to change notification settings - Fork 0
/
test_fca_oo.py
202 lines (166 loc) · 6.75 KB
/
test_fca_oo.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
# -*- coding: utf-8 -*-
"""
Created on Mon Jul 11 10:28:55 2016
@author: ator-
To run right click project and open command prompt.
Type py.test.
"""
import pytest
from hypothesis import given
import hypothesis.strategies as st
import hypothesis.extra.numpy as st_np
import numpy as np
import pandas as pd
import kernel_fca_oo as krn
@pytest.fixture(scope='function')
def kernel_classes():
return [krn.KernelSystemNP, krn.KernelSystemDF, krn.FCASystemDF]
arr_shape_strat = st.tuples(st.integers(min_value=2, max_value=5),
st.integers(min_value=2, max_value=5))
bin_int_strat = st.integers(min_value=0, max_value=1)
arr_strat = arr_shape_strat.flatmap(lambda t: st_np.arrays(np.int8, (t[0], t[1]),
elements=bin_int_strat))
intent_strat = st.shared(arr_strat, key=1).flatmap(lambda a: st.sets(st.integers(0, a.shape[1]-1)))
@given(st.shared(arr_strat, key=1), intent_strat)
def test_ei(kernel_classes, arr, gen_intent):
#print(arr)
#print(arr2)
#print('_______')
df = pd.DataFrame(arr)
for K in kernel_classes:
if K == krn.KernelSystemNP:
ks = K(arr)
else:
ks = K(df)
extent = ks.extent(gen_intent)
intent = set(ks.intent(extent))
assert intent >= gen_intent
extent2 = ks.extent(intent)
assert extent == extent2
# assert len(arr) > 9
@given(arr_strat)
def test_mf(kernel_classes, arr):
df = pd.DataFrame(arr)
for K in kernel_classes:
if K == krn.KernelSystemNP:
ks = K(arr)
else:
ks = K(df)
mf = ks.minusframe()
assert len(mf) == arr.shape[0], 'error in ' + K.__name__
for i in mf.index:
row = mf.loc[i]
assert row['w'] >= np.sum(arr[i]), 'error in ' + K.__name__
assert row['w'] <= np.sum(arr[i])*len(arr), 'error in ' + K.__name__
def test_cc():
bin_slr = np.array([[0, 0, 1, 1, 1, 1, 0],
[0, 0, 0, 1, 1, 1, 1],
[1, 0, 1, 0, 1, 1, 1],
[1, 0, 1, 1, 1, 0, 0],
[1, 1, 1, 0, 1, 0, 0],
[1, 1, 1, 1, 1, 0, 0],
[1, 1, 0, 0, 0, 0, 0],
[1, 1, 1, 0, 0, 0, 0]])
bin_three_aspects = np.array([[1, 1, 1, 1, 0, 0, 0, 0],
[1, 1, 0, 0, 1, 1, 1, 0],
[1, 1, 0, 0, 0, 0, 0, 1],
[1, 0, 0, 0, 0, 0, 0, 0],
[1, 0, 0, 0, 0, 0, 0, 0],
[0, 1, 0, 0, 1, 1, 1, 0],
[0, 0, 0, 0, 1, 1, 1, 0],
[0, 0, 0, 0, 1, 1, 1, 0],
[0, 0, 0, 0, 1, 1, 1, 0]])
for System in [krn.FCASystemDF,
krn.FCAPathSystemDF,
krn.FreqLexiSeriateSystem,
krn.ConfLexiSeriateSystem,
krn.LexiSystem,
krn.LexiTSystem,
lambda data: krn.FreqLexiSeriateSystem(data, refill=True),
lambda data: krn.ConfLexiSeriateSystem(data, refill=True)]:
ks = System(pd.DataFrame(bin_slr))
ccc = ks.conceptchaincover()
assert ccc[0][0].area() > 15
# ks = System(bin_slr)
if not isinstance(ks, krn.LexiSystem):
ccc2 = ks.conceptchaincover_v2()
assert ccc2[0][0].area() > 15
ks2 = System(pd.DataFrame(bin_three_aspects))
chainlist, _ = ks2.conceptchaincover(uncovered=0.0)
assert 1 < len(chainlist) < 6 # Ideally 3 chains
ks = krn.FCAPathSystemDF(pd.DataFrame(bin_slr))
cr = ks.conceptrec([])
assert cr.intent == set()
assert cr.extent == set(range(8))
cr1 = ks.conceptrec([1])
assert ks.conceptdist(cr, cr1) == 4*2
for System in [krn.LexiSystem, krn.LexiTSystem]:
ks = System(pd.DataFrame(bin_three_aspects))
ccc, _ = ks.conceptchaincover(uncovered=0.0, min_cost=True)
assert 0 < len(ccc) < 3 # Test cost minimization
@given(st_np.arrays(np.int8, (5, 4), elements=bin_int_strat))
def test_l2_cover_superiority(arr):
data = pd.DataFrame(arr)
ls = krn.LexiSystem(data, full_lexi=False)
lsf = krn.LexiSystem(data, full_lexi=True)
ccc, _ = ls.conceptchaincover(uncovered=0.0)
ccc_f, _ = lsf.conceptchaincover(uncovered=0.0)
assert len(ccc_f) <= len(ccc) #Maybe random superiority?
@given(st_np.arrays(np.int8, (8, 6), elements=bin_int_strat),
st_np.arrays(np.int8, (8, 6), elements=bin_int_strat))
def test_transpose_system_cover_superiority(data, arr_base):
arr = pd.DataFrame(data * arr_base) # arr 1-s will be subset of data
data = pd.DataFrame(data)
ls = krn.LexiSystem(data, full_lexi=False)
lst = krn.LexiTSystem(data, full_lexi=False)
cc = ls.chain_from_array(arr)
cct = lst.chain_from_array(arr)
assert cc.cover(arr) <= cct.cover(arr)
def test_ConceptChain():
bin_slr = np.array([[0, 0, 1, 1, 1, 1, 0],
[0, 0, 0, 1, 1, 1, 1],
[1, 0, 1, 0, 1, 1, 1],
[1, 0, 1, 1, 1, 0, 0],
[1, 1, 1, 0, 1, 0, 0],
[1, 1, 1, 1, 1, 0, 0],
[1, 1, 0, 0, 0, 0, 0],
[1, 1, 1, 0, 0, 0, 0]])
ks = krn.FCASystemDF(pd.DataFrame(bin_slr))
cc = krn.ConceptChain([1, 2, 3, 6], ks)
assert len(cc) == 3
assert [(len(e), len(i)) for e, i in cc] == [(1,4), (2,3), (6,1)]
ccT = krn.ConceptChain.intent_init([1, 2, 3, 6], ks)
assert len(ccT) == 3
assert [(len(e), len(i)) for e, i in ccT] == [(1,5), (3,3), (4,2)]
#assert ccT.intent_labels() == [1, 2, 3, 6]
df_bin_slr = pd.DataFrame(bin_slr.copy())
for a in [df_bin_slr, bin_slr]:
expected = 11
assert cc.cover(a) == expected
for r, c in [(2, 5), (1, 4), (1, 3), (2, 4), (3, 4), (4, 4), (5, 4)]:
try:
a.loc[r, c] = 0 # Dataframe
except:
a[r, c] = 0 #array
expected = expected - 1
assert cc.cover(a) == expected
# =============================================================================
# @given(st.integers(), st.integers())
# def test_ints_are_commutative(x, y):
# assert x + y == y + x
# assert x * y == y * x
# #assert x / y == 1 / (y / x)
#
# @pytest.fixture(scope='function')
# def some_list():
# return list(range(5))
#
# def test_answer():
# assert 5 == 5
#
# def test_answer2(some_list):
# assert len(some_list) == 5
#
# =============================================================================
if __name__ == "__main__":
pytest.main()