Skip to content

Commit 99613b4

Browse files
rhshadrachrhshadrach
committed
WIP
1 parent 02708ed commit 99613b4

20 files changed

+64
-42
lines changed

‎pandas/core/construction.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -652,8 +652,8 @@ def sanitize_array(
652652
data = list(data)
653653

654654
if len(data) == 0 and dtype is None:
655-
# We default to float64, matching numpy
656-
subarr = np.array([], dtype=np.float64)
655+
# We default to object, diverging from NumPy
656+
subarr = np.array([], dtype=np.object_)
657657

658658
elif dtype is not None:
659659
subarr = _try_cast(data, dtype, copy)

‎pandas/core/groupby/generic.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -578,7 +578,7 @@ def _transform_general(
578578
concatenated = concat(results, ignore_index=True)
579579
result = self._set_result_index_ordered(concatenated)
580580
else:
581-
result = self.obj._constructor(dtype=np.float64)
581+
result = self.obj._constructor(dtype=self.obj.dtype)
582582

583583
result.name = self.obj.name
584584
return result

‎pandas/core/internals/managers.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1778,7 +1778,7 @@ def as_array(
17781778
passed_nan = lib.is_float(na_value) and isna(na_value)
17791779

17801780
if len(self.blocks) == 0:
1781-
arr = np.empty(self.shape, dtype=float)
1781+
arr = np.empty(self.shape, dtype=object)
17821782
return arr.transpose()
17831783

17841784
if self.is_single_block:

‎pandas/tests/arrays/categorical/test_missing.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -122,9 +122,9 @@ def test_compare_categorical_with_missing(self, a1, a2, categories):
122122
"na_value, dtype",
123123
[
124124
(pd.NaT, "datetime64[ns]"),
125-
(None, "float64"),
125+
(None, "object"),
126126
(np.nan, "float64"),
127-
(pd.NA, "float64"),
127+
(pd.NA, "object"),
128128
],
129129
)
130130
def test_categorical_only_missing_values_no_cast(self, na_value, dtype):

‎pandas/tests/frame/methods/test_quantile.py

Lines changed: 9 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -81,7 +81,7 @@ def test_quantile(self, datetime_frame, interp_method, request):
8181
def test_empty(self, interp_method):
8282
interpolation, method = interp_method
8383
q = DataFrame({"x": [], "y": []}).quantile(
84-
0.1, axis=0, numeric_only=True, interpolation=interpolation, method=method
84+
0.1, axis=0, interpolation=interpolation, method=method
8585
)
8686
assert np.isnan(q["x"]) and np.isnan(q["y"])
8787

@@ -319,8 +319,11 @@ def test_quantile_multi_empty(self, interp_method):
319319
result = DataFrame({"x": [], "y": []}).quantile(
320320
[0.1, 0.9], axis=0, interpolation=interpolation, method=method
321321
)
322+
dtype = "float64" if interpolation == "linear" else "object"
322323
expected = DataFrame(
323-
{"x": [np.nan, np.nan], "y": [np.nan, np.nan]}, index=[0.1, 0.9]
324+
{"x": [np.nan, np.nan], "y": [np.nan, np.nan]},
325+
index=[0.1, 0.9],
326+
dtype=dtype,
324327
)
325328
tm.assert_frame_equal(result, expected)
326329

@@ -655,7 +658,8 @@ def test_quantile_empty_no_rows_floats(self, interp_method):
655658
tm.assert_frame_equal(res, exp)
656659

657660
res = df.quantile(0.5, axis=1, interpolation=interpolation, method=method)
658-
exp = Series([], index=[], dtype="float64", name=0.5)
661+
dtype = "float64" if interpolation == "linear" else "object"
662+
exp = Series([], index=[], name=0.5, dtype=dtype)
659663
tm.assert_series_equal(res, exp)
660664

661665
res = df.quantile([0.5], axis=1, interpolation=interpolation, method=method)
@@ -710,7 +714,8 @@ def test_quantile_empty_no_columns(self, interp_method):
710714
result = df.quantile(
711715
0.5, numeric_only=True, interpolation=interpolation, method=method
712716
)
713-
expected = Series([], index=[], name=0.5, dtype=np.float64)
717+
dtype = "float64" if interpolation == "linear" else "object"
718+
expected = Series([], index=[], name=0.5, dtype=dtype)
714719
expected.index.name = "captain tightpants"
715720
tm.assert_series_equal(result, expected)
716721

‎pandas/tests/frame/methods/test_reindex.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -77,7 +77,7 @@ def test_setitem_reset_index_dtypes(self):
7777
df1["d"] = []
7878
result = df1.reset_index()
7979
expected = DataFrame(columns=["a", "b", "c", "d"], index=range(0)).astype(
80-
{"a": "datetime64[ns]", "b": np.int64, "c": np.float64, "d": np.float64}
80+
{"a": "datetime64[ns]", "b": np.int64, "c": np.float64, "d": np.object_}
8181
)
8282
tm.assert_frame_equal(result, expected)
8383

‎pandas/tests/frame/test_reductions.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1627,7 +1627,7 @@ def test_min_max_dt64_api_consistency_empty_df(self):
16271627
# check DataFrame/Series api consistency when calling min/max on an empty
16281628
# DataFrame/Series.
16291629
df = DataFrame({"x": []})
1630-
expected_float_series = Series([], dtype=float)
1630+
expected_float_series = Series([], dtype=object)
16311631
# check axis 0
16321632
assert np.isnan(df.min(axis=0).x) == np.isnan(expected_float_series.min())
16331633
assert np.isnan(df.max(axis=0).x) == np.isnan(expected_float_series.max())

‎pandas/tests/frame/test_stack_unstack.py

Lines changed: 10 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -1418,11 +1418,12 @@ def test_stack_timezone_aware_values(future_stack):
14181418
def test_stack_empty_frame(dropna, future_stack):
14191419
# GH 36113
14201420
levels = [np.array([], dtype=np.int64), np.array([], dtype=np.int64)]
1421-
expected = Series(dtype=np.float64, index=MultiIndex(levels=levels, codes=[[], []]))
1421+
expected = Series(dtype=np.object_, index=MultiIndex(levels=levels, codes=[[], []]))
14221422
if future_stack and dropna is not lib.no_default:
14231423
with pytest.raises(ValueError, match="dropna must be unspecified"):
14241424
DataFrame(dtype=np.float64).stack(dropna=dropna, future_stack=future_stack)
14251425
else:
1426+
# dtype=np.float64 is lost since there are no columns
14261427
result = DataFrame(dtype=np.float64).stack(
14271428
dropna=dropna, future_stack=future_stack
14281429
)
@@ -1612,7 +1613,9 @@ def test_unstack(self, multiindex_year_month_day_dataframe_random_data):
16121613
(
16131614
[[1, 1, None, None, 30.0], [2, None, None, None, 30.0]],
16141615
["ix1", "ix2", "col1", "col2", "col3"],
1615-
None,
1616+
# Nones are used as floats in the presence of numeric data,
1617+
# resulting in np.nan for index level 1.
1618+
np.nan,
16161619
[None, None, 30.0],
16171620
),
16181621
],
@@ -1624,10 +1627,12 @@ def test_unstack_partial(
16241627
# https://github.com/pandas-dev/pandas/issues/19351
16251628
# make sure DataFrame.unstack() works when its run on a subset of the DataFrame
16261629
# and the Index levels contain values that are not present in the subset
1627-
result = DataFrame(result_rows, columns=result_columns).set_index(
1628-
["ix1", "ix2"]
1630+
data = (
1631+
DataFrame(result_rows, columns=result_columns)
1632+
.set_index(["ix1", "ix2"])
1633+
.iloc[1:2]
16291634
)
1630-
result = result.iloc[1:2].unstack("ix2")
1635+
result = data.unstack("ix2")
16311636
expected = DataFrame(
16321637
[expected_row],
16331638
columns=MultiIndex.from_product(

‎pandas/tests/groupby/methods/test_quantile.py

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -192,7 +192,7 @@ def test_quantile_missing_group_values_no_segfaults():
192192
([1.0, np.nan, 2.0, 2.0], range(4), [1.0, 2.0], [0.0, 2.5]),
193193
(["a", "b", "b", np.nan], range(4), ["a", "b"], [0, 1.5]),
194194
([0], [42], [0], [42.0]),
195-
([], [], np.array([], dtype="float64"), np.array([], dtype="float64")),
195+
([], np.array([], dtype="float64"), [], np.array([], dtype="float64")),
196196
],
197197
)
198198
def test_quantile_missing_group_values_correct_results(

‎pandas/tests/groupby/test_apply.py

Lines changed: 1 addition & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1479,9 +1479,7 @@ def test_empty_df(method, op):
14791479
group = getattr(gb, "b")
14801480

14811481
result = getattr(group, method)(op)
1482-
expected = Series(
1483-
[], name="b", dtype="float64", index=Index([], dtype="float64", name="a")
1484-
)
1482+
expected = Series([], name="b", index=Index([], name="a"))
14851483

14861484
tm.assert_series_equal(result, expected)
14871485

0 commit comments

Comments
 (0)