update unit test

erdogant · Oct 30, 2024 · d37cfc8 · d37cfc8
1 parent eb4edde
commit d37cfc8
Showing 1 changed file with 55 additions and 43 deletions.
diff --git a/benfordslaw/tests/test_benfordslaw.py b/benfordslaw/tests/test_benfordslaw.py
@@ -6,46 +6,58 @@
 import unittest
 
 class TestBENFORDSLAW(unittest.TestCase):
-	def test_nums(self):
-		bl = benfordslaw()
-		X = [12, 23, 34, 45, 56, 67, 78, 89, 90, 100]
-		out = bl.fit(np.array(X))
-		np.all(out['percentage_emp'][:,1]==[20, 10, 10, 10, 10, 10, 10, 10, 10])
-
-	def test_benfordslaw(self):
-		bl = benfordslaw()
-		df = bl.import_example()
-		Iloc = df['candidate']=='Donald Trump'
-		X = df['votes'].loc[Iloc].values
-
-		# TEST 1: check output is unchanged
-		methods=[None, 'chi2', 'ks']
-		for method in methods:
-			bl = benfordslaw(method=method)
-			out = bl.fit(X)
-			assert np.all(np.isin([*out.keys()], ['P', 't', 'percentage_emp', 'P_significant']))
-
-		# TEST 2: check chi2
-		bl = benfordslaw(method='chi2')
-		out = bl.fit(X)
-		assert out['P']==0.4150723856863902
-		assert out['t']==8.190646662956185
-		assert out['P_significant']==False
-
-		# TEST 3: check ks
-		bl = benfordslaw(method='ks')
-		out = bl.fit(X)
-		assert out['P']==1.0
-		assert out['t']==0.1111111111111111
-
-		# TEST 4: check percentages
-		assert np.all(out['percentage_emp'][:,1]==[31.64521544487969, 16.508114157806382, 12.479015109121432, 9.820928931169558, 7.862339115836598, 6.3794068270845, 5.8477895914941245, 4.616675993284835, 4.8405148293228875])
-
-	def test_dataframe(self):
-		bl = benfordslaw()
-		df = bl.import_example()
-		Iloc = df['candidate']=='Donald Trump'
-		X = df['votes'].loc[Iloc]
-		results = bl.fit(X)
-		np.all(np.isin([*results.keys()], ['P', 't', 'percentage_emp', 'P_significant']))
-		assert np.all(results['percentage_emp'][:,1]==[31.64521544487969, 16.508114157806382, 12.479015109121432, 9.820928931169558, 7.862339115836598, 6.3794068270845, 5.8477895914941245, 4.616675993284835, 4.8405148293228875])
+    def test_benfordslaw_digits(self):
+        gt = {-6: True, -5: True, -4: True, -3: True, -2: True, -1: True, 1: True, 2: True, 3: True, 4: True, 5: True, 6: True}
+        total_results = []
+        pos = [-9, -8, -7, -6, -5, -4, -3, -2 -1, 1, 2, 3, 4, 5, 6, 7, 8, 9]
+        for p, t in zip(pos, gt):
+            bl = benfordslaw(pos=p, verbose=0)
+            df = bl.import_example(data='elections_usa', verbose=0)
+            Iloc = df['candidate']=='Donald Trump'
+            X = df['votes'].loc[Iloc].values
+            results = bl.fit(X)
+            assert gt.get(p, None)!=np.isnan(results['P'])
+
+    def test_nums(self):
+        bl = benfordslaw()
+        X = [12, 23, 34, 45, 56, 67, 78, 89, 90, 100]
+        out = bl.fit(np.array(X))
+        np.all(out['percentage_emp'][:,1]==[20, 10, 10, 10, 10, 10, 10, 10, 10])
+
+    def test_benfordslaw(self):
+        bl = benfordslaw()
+        df = bl.import_example()
+        Iloc = df['candidate']=='Donald Trump'
+        X = df['votes'].loc[Iloc].values
+
+        # TEST 1: check output is unchanged
+        methods=[None, 'chi2', 'ks']
+        for method in methods:
+            bl = benfordslaw(method=method)
+            out = bl.fit(X)
+            assert np.all(np.isin([*out.keys()], ['P', 't', 'percentage_emp', 'P_significant']))
+
+        # TEST 2: check chi2
+        bl = benfordslaw(method='chi2')
+        out = bl.fit(X)
+        assert out['P']==0.4150723856863902
+        assert out['t']==8.190646662956185
+        assert out['P_significant']==False
+
+        # TEST 3: check ks
+        bl = benfordslaw(method='ks')
+        out = bl.fit(X)
+        assert out['P']==1.0
+        assert out['t']==0.1111111111111111
+
+        # TEST 4: check percentages
+        assert np.all(out['percentage_emp'][:,1]==[31.64521544487969, 16.508114157806382, 12.479015109121432, 9.820928931169558, 7.862339115836598, 6.3794068270845, 5.8477895914941245, 4.616675993284835, 4.8405148293228875])
+
+    def test_dataframe(self):
+        bl = benfordslaw()
+        df = bl.import_example()
+        Iloc = df['candidate']=='Donald Trump'
+        X = df['votes'].loc[Iloc]
+        results = bl.fit(X)
+        np.all(np.isin([*results.keys()], ['P', 't', 'percentage_emp', 'P_significant']))
+        assert np.all(results['percentage_emp'][:,1]==[31.64521544487969, 16.508114157806382, 12.479015109121432, 9.820928931169558, 7.862339115836598, 6.3794068270845, 5.8477895914941245, 4.616675993284835, 4.8405148293228875])