Skip to content

Commit 2ec380a

Browse files
wali-rehemanwali-reheman
committed
fix tests: add mutual_information column to existing test DataFrames, add new test
1 parent 9a307bc commit 2ec380a

1 file changed

Lines changed: 70 additions & 10 deletions

File tree

‎mlxtend/frequent_patterns/tests/test_association_rules.py‎

Lines changed: 70 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -50,6 +50,7 @@
5050
"jaccard",
5151
"certainty",
5252
"kulczynski",
53+
"mutual_information",
5354
]
5455

5556

@@ -65,15 +66,15 @@ def test_default():
6566

6667
expect = pd.DataFrame(
6768
[
68-
[(8,), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8],
69-
[(6,), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8],
70-
[(8, 3), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8],
71-
[(8, 5), (3,), 0.6, 0.8, 0.6, 1.0, 1.25, 1.0, 0.12, np.inf, 0.5, 0.75, 1.0, 0.875],
72-
[(8,), (3, 5), 0.6, 0.8, 0.6, 1.0, 1.25, 1.0, 0.12, np.inf, 0.5, 0.75, 1.0, 0.875],
73-
[(3,), (5,), 0.8, 1.0, 0.8, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.8, 0.0, 0.9],
74-
[(5,), (3,), 1.0, 0.8, 0.8, 0.8, 1.0, 1.0, 0.0, 1.0, 0.0, 0.8, 0.0, 0.9],
75-
[(10,), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8],
76-
[(8,), (3,), 0.6, 0.8, 0.6, 1.0, 1.25, 1.0, 0.12, np.inf, 0.5, 0.75, 1.0, 0.875],
69+
[(8,), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8, 0.0],
70+
[(6,), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8, 0.0],
71+
[(8, 3), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8, 0.0],
72+
[(8, 5), (3,), 0.6, 0.8, 0.6, 1.0, 1.25, 1.0, 0.12, np.inf, 0.5, 0.75, 1.0, 0.875, 0.0],
73+
[(8,), (3, 5), 0.6, 0.8, 0.6, 1.0, 1.25, 1.0, 0.12, np.inf, 0.5, 0.75, 1.0, 0.875, 0.0],
74+
[(3,), (5,), 0.8, 1.0, 0.8, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.8, 0.0, 0.9, 0.0],
75+
[(5,), (3,), 1.0, 0.8, 0.8, 0.8, 1.0, 1.0, 0.0, 1.0, 0.0, 0.8, 0.0, 0.9, 0.0],
76+
[(10,), (5,), 0.6, 1.0, 0.6, 1.0, 1.0, 1.0, 0.0, np.inf, 0, 0.6, 0.0, 0.8, 0.0],
77+
[(8,), (3,), 0.6, 0.8, 0.6, 1.0, 1.25, 1.0, 0.12, np.inf, 0.5, 0.75, 1.0, 0.875, 0.0],
7778
],
7879

7980
columns=columns_ordered,
@@ -120,6 +121,7 @@ def test_nullability():
120121
0.667,
121122
0,
122123
0.833,
124+
0.0,
123125
],
124126
[
125127
(10, 5),
@@ -136,6 +138,7 @@ def test_nullability():
136138
0.667,
137139
0.0,
138140
0.833,
141+
0.0,
139142
],
140143
[
141144
(10,),
@@ -152,6 +155,7 @@ def test_nullability():
152155
0.615,
153156
0.0,
154157
0.833,
158+
0.415,
155159
],
156160
[
157161
(10,),
@@ -168,6 +172,7 @@ def test_nullability():
168172
0.615,
169173
0.0,
170174
0.833,
175+
0.415,
171176
],
172177
[
173178
(10,),
@@ -184,6 +189,7 @@ def test_nullability():
184189
0.615,
185190
0,
186191
0.833,
192+
-0.169,
187193
],
188194
[
189195
(3, 5),
@@ -200,6 +206,7 @@ def test_nullability():
200206
0.615,
201207
-0.333,
202208
0.833,
209+
-0.169,
203210
],
204211
[
205212
(3,),
@@ -216,6 +223,7 @@ def test_nullability():
216223
0.667,
217224
0.0,
218225
0.833,
226+
0.0,
219227
],
220228
[
221229
(3,),
@@ -232,6 +240,7 @@ def test_nullability():
232240
0.615,
233241
-0.333,
234242
0.833,
243+
-0.169,
235244
],
236245
[(3,), (5,), 1.0, 1.0, 1.0, 1.0, 1.0, 0.8, 0.0, np.inf, 0, 1.0, 0, 1.0],
237246
[
@@ -249,6 +258,7 @@ def test_nullability():
249258
0.667,
250259
0,
251260
0.833,
261+
0.0,
252262
],
253263
[
254264
(5,),
@@ -265,8 +275,25 @@ def test_nullability():
265275
0.615,
266276
-0.333,
267277
0.833,
278+
-0.169,
279+
],
280+
[
281+
(5,),
282+
(3,),
283+
1.0,
284+
1.0,
285+
1.0,
286+
1.0,
287+
1.0,
288+
0.8,
289+
0.0,
290+
np.inf,
291+
0,
292+
1.0,
293+
0.0,
294+
1.0,
295+
0.0,
268296
],
269-
[(5,), (3,), 1.0, 1.0, 1.0, 1.0, 1.0, 0.8, 0.0, np.inf, 0, 1.0, 0.0, 1.0],
270297
],
271298
columns=columns_ordered,
272299
)
@@ -335,6 +362,7 @@ def test_empty_result():
335362
"jaccard",
336363
"certainty",
337364
"kulczynski",
365+
"mutual_information",
338366
]
339367
)
340368
res_df = association_rules(df_freq_items, len(df), min_threshold=2)
@@ -563,3 +591,35 @@ def test_mutual_information_metric():
563591

564592
# Non-existent itemsets (support=0) should give -inf
565593
assert res_df["mutual_information"].notna().any()
594+
595+
596+
def test_mutual_information_metric():
597+
"""Test mutual_information metric returns correct values."""
598+
import math
599+
600+
res_df = association_rules(
601+
df_freq_items_with_colnames,
602+
len(df),
603+
metric="mutual_information",
604+
min_threshold=-100,
605+
)
606+
assert "mutual_information" in res_df.columns
607+
608+
# Eggs -> Kidney Beans: sAC=0.6, sA=0.8, sC=1.0
609+
# MI = log2(0.6 / (0.8 * 1.0)) = log2(0.75) approx -0.415
610+
rule = res_df[
611+
res_df["antecedents"].apply(lambda x: x == frozenset({"Eggs"}))
612+
& res_df["consequents"].apply(lambda x: x == frozenset({"Kidney Beans"}))
613+
]
614+
assert len(rule) == 1
615+
expected_mi = math.log2(0.6 / (0.8 * 1.0))
616+
assert abs(rule["mutual_information"].values[0] - expected_mi) < 1e-6
617+
618+
# Milk -> Kidney Beans: sAC=0.6, sA=0.6, sC=1.0
619+
# MI = log2(0.6 / (0.6 * 1.0)) = log2(1) = 0
620+
milk_rule = res_df[
621+
res_df["antecedents"].apply(lambda x: x == frozenset({"Milk"}))
622+
& res_df["consequents"].apply(lambda x: x == frozenset({"Kidney Beans"}))
623+
]
624+
assert len(milk_rule) == 1
625+
assert abs(milk_rule["mutual_information"].values[0]) < 1e-6

0 commit comments

Comments
 (0)