Skip to content

TST: refactored test_factorize #32311

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Mar 7, 2020
Merged

TST: refactored test_factorize #32311

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
e419141
refactored test_factorize
SaturnFromTitan Feb 27, 2020
cdca771
fixing ci
SaturnFromTitan Feb 27, 2020
1b4825b
switched to using intp instead of workaround to make CI happy
SaturnFromTitan Feb 27, 2020
9c22a50
review comments
SaturnFromTitan Mar 4, 2020
d26d0fa
Merge branch 'master' into fixturize-test_factorize
SaturnFromTitan Mar 4, 2020
76d73cc
Merge branch 'master' into fixturize-test_factorize
SaturnFromTitan Mar 5, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
73 changes: 17 additions & 56 deletions pandas/tests/base/test_ops.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -533,66 +533,27 @@ def test_value_counts_datetime64(self, index_or_series):
result2 = td2.value_counts()
tm.assert_series_equal(result2, expected_s)

def test_factorize(self):
for orig in self.objs:
o = orig.copy()

if isinstance(o, Index) and o.is_boolean():
exp_arr = np.array([0, 1] + [0] * 8, dtype=np.intp)
exp_uniques = o
exp_uniques = Index([False, True])
else:
exp_arr = np.array(range(len(o)), dtype=np.intp)
exp_uniques = o
codes, uniques = o.factorize()

tm.assert_numpy_array_equal(codes, exp_arr)
if isinstance(o, Series):
tm.assert_index_equal(uniques, Index(orig), check_names=False)
else:
# factorize explicitly resets name
tm.assert_index_equal(uniques, exp_uniques, check_names=False)

def test_factorize_repeated(self):
for orig in self.objs:
o = orig.copy()
@pytest.mark.parametrize("sort", [True, False])
def test_factorize(self, index_or_series_obj, sort):
obj = index_or_series_obj
result_codes, result_uniques = obj.factorize(sort=sort)

# don't test boolean
if isinstance(o, Index) and o.is_boolean():
continue
constructor = pd.Index
if isinstance(obj, pd.MultiIndex):
constructor = pd.MultiIndex.from_tuples
expected_uniques = constructor(obj.unique())

# sort by value, and create duplicates
if isinstance(o, Series):
o = o.sort_values()
n = o.iloc[5:].append(o)
else:
indexer = o.argsort()
o = o.take(indexer)
n = o[5:].append(o)

exp_arr = np.array(
[5, 6, 7, 8, 9, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype=np.intp
)
codes, uniques = n.factorize(sort=True)

tm.assert_numpy_array_equal(codes, exp_arr)
if isinstance(o, Series):
tm.assert_index_equal(
uniques, Index(orig).sort_values(), check_names=False
)
else:
tm.assert_index_equal(uniques, o, check_names=False)
if sort:
expected_uniques = expected_uniques.sort_values()

exp_arr = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 2, 3, 4], np.intp)
codes, uniques = n.factorize(sort=False)
tm.assert_numpy_array_equal(codes, exp_arr)
# construct an integer ndarray so that
# `expected_uniques.take(expected_codes)` is equal to `obj`
expected_uniques_list = list(expected_uniques)
expected_codes = [expected_uniques_list.index(val) for val in obj]
expected_codes = np.asarray(expected_codes, dtype=np.intp)
Comment on lines +552 to +553
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you not just use np.take here instead?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

you can use expected_uniques.take is better

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't think I can use take here. I want to construct an array containing the indices in expected_uniques of the values of obj.

# given
obj = Series([1, 2, 1, 3, 5])
expected_uniques = obj.unique()  # array([1, 2, 3, 5])

# needed
expected_codes = array([0, 1, 0, 2, 3])

I could only use take if already have the indices and need the values. I basically need the reverse of take.

I guess I could use where somehow, but it will probably be more complex than just using vanilla python list.index()

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think you can just use pd.factorize then

In [7]: import pandas as pd
In [8]: obj = pd.Series([1, 2, 1, 3, 5])
In [10]: pd.factorize(obj)
Out[10]: (array([0, 1, 0, 2, 3]), Int64Index([1, 2, 3, 5], dtype='int64'))

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm testing factorize here, so I need an alternative implementation 😄

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok you can actually use .get_loc but a simple impl is better; can you add a comment explaining what you are doing (factorizing)


if isinstance(o, Series):
expected = Index(o.iloc[5:10].append(o.iloc[:5]))
tm.assert_index_equal(uniques, expected, check_names=False)
else:
expected = o[5:10].append(o[:5])
tm.assert_index_equal(uniques, expected, check_names=False)
tm.assert_numpy_array_equal(result_codes, expected_codes)
tm.assert_index_equal(result_uniques, expected_uniques)

def test_duplicated_drop_duplicates_index(self):
# GH 4060
Expand Down