def test_run_1(self):
# should have the same output as test_run_2, but _do_multiprocessing() is a MagicMock
in_a = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_b = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_c = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_series = [in_a, in_b, in_c]
expected = DataFrame(
{
0: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
1: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
2: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
u"all": Series({1: 15, 2: 6, 3: 6, 4: 3, 5: 3}),
}
)
exp = FrequencyExperimenter(in_series)
exp._do_multiprocessing = mock.MagicMock()
exp._do_multiprocessing.return_value = [
(0, Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1})),
(1, Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1})),
(2, Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1})),
]
actual = exp.run()
exp._do_multiprocessing.assert_called_once_with(experimenter_func, [[(0, in_a)], [(1, in_b)], [(2, in_c)]])
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected.loc[:, i].index), list(actual.loc[:, i].index))
self.assertSequenceEqual(list(expected.loc[:, i].values), list(actual.loc[:, i].values))
def test_run_2(self):
# should have the same output as test_run_1, but without the MagicMock
in_a = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_b = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_c = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_series = [in_a, in_b, in_c]
expected = DataFrame(
{
0: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
1: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
2: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
u"all": Series({1: 15, 2: 6, 3: 6, 4: 3, 5: 3}),
}
)
exp = FrequencyExperimenter(in_series)
actual = exp.run()
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected.loc[:, i].index), list(actual.loc[:, i].index))
self.assertSequenceEqual(list(expected.loc[:, i].values), list(actual.loc[:, i].values))
def test_run_4(self):
# same as test_run_3, but input is a dict
in_a = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_b = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1, 4, 4, 3, 5, 1, 1, 1])
in_c = Series([1, 2, 1, 1, 3, 1, 3, 2, 1])
in_series = {u"hello": in_a, u"zello": in_b, u"jello": in_c}
expected = DataFrame(
{
u"hello": Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
u"zello": Series({1: 8, 2: 2, 3: 3, 4: 3, 5: 2}),
u"jello": Series({1: 5, 2: 2, 3: 2}),
u"all": Series({1: 18, 2: 6, 3: 7, 4: 4, 5: 3}),
}
)
exp = FrequencyExperimenter(in_series)
actual = exp.run()
# because numpy's NaN != NaN
actual = actual.fillna(value=4000)
expected = expected.fillna(value=4000)
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected.loc[:, i].index), list(actual.loc[:, i].index))
self.assertSequenceEqual(list(expected.loc[:, i].values), list(actual.loc[:, i].values))
def test_run_3(self):
# more complicated arithmetic
in_a = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1])
in_b = Series([1, 2, 1, 1, 3, 1, 4, 5, 3, 2, 1, 4, 4, 3, 5, 1, 1, 1])
in_c = Series([1, 2, 1, 1, 3, 1, 3, 2, 1])
in_series = [in_a, in_b, in_c]
expected = DataFrame(
{
0: Series({1: 5, 2: 2, 3: 2, 4: 1, 5: 1}),
1: Series({1: 8, 2: 2, 3: 3, 4: 3, 5: 2}),
2: Series({1: 5, 2: 2, 3: 2}),
u"all": Series({1: 18, 2: 6, 3: 7, 4: 4, 5: 3}),
}
)
exp = FrequencyExperimenter(in_series)
actual = exp.run()
# because numpy's NaN != NaN
actual = actual.fillna(value=4000)
expected = expected.fillna(value=4000)
self.assertEqual(len(expected.columns), len(actual.columns))
for i in expected.columns:
self.assertSequenceEqual(list(expected.loc[:, i].index), list(actual.loc[:, i].index))
self.assertSequenceEqual(list(expected.loc[:, i].values), list(actual.loc[:, i].values))
