def test_collapse_missing_table_ids_in_taxonomy(self):
table = pd.DataFrame([[2.0, 2.0], [1.0, 1.0], [9.0, 8.0], [0.0, 4.0]],
index=['A', 'B', 'C', 'D'],
columns=['feat1', 'feat2'])
taxonomy = pd.Series(['a; b; c', 'a; b; d'], index=['feat1', 'feat3'])
with self.assertRaisesRegex(ValueError, 'missing.*feat2'):
collapse(table, taxonomy, 1)
def test_collapse_bad_level(self):
table = pd.DataFrame([[2.0, 2.0], [1.0, 1.0], [9.0, 8.0], [0.0, 4.0]],
index=['A', 'B', 'C', 'D'],
columns=['feat1', 'feat2'])
taxonomy = pd.Series(['a; b; c', 'a; b; d'], index=['feat1', 'feat2'])
with self.assertRaisesRegex(ValueError, 'of 42 is larger'):
collapse(table, taxonomy, 42)
with self.assertRaisesRegex(ValueError, 'of 0 is too low'):
collapse(table, taxonomy, 0)
def test_collapse_missing_level(self):
table = pd.DataFrame([[2.0, 2.0], [1.0, 1.0], [9.0, 8.0], [0.0, 4.0]],
index=['A', 'B', 'C', 'D'],
columns=['feat1', 'feat2'])
taxonomy = pd.Series(['a; b', 'a; b; d'], index=['feat1', 'feat2'])
actual = collapse(table, taxonomy, 1)
expected = pd.DataFrame([[4.0], [2.0], [17.0], [4.0]],
index=['A', 'B', 'C', 'D'],
columns=['a'])
self.assert_data_frame_almost_equal(actual, expected)
actual = collapse(table, taxonomy, 2)
expected = pd.DataFrame([[4.0], [2.0], [17.0], [4.0]],
index=['A', 'B', 'C', 'D'],
columns=['a;b'])
self.assert_data_frame_almost_equal(actual, expected)
actual = collapse(table, taxonomy, 3)
expected = pd.DataFrame(
[[2.0, 2.0], [1.0, 1.0], [9.0, 8.0], [0.0, 4.0]],
index=['A', 'B', 'C', 'D'],
columns=['a;b;__', 'a;b;d'])
self.assert_data_frame_almost_equal(actual, expected)