+253
-274
lines changedFilter options
+253
-274
lines changed Original file line number Diff line number Diff line change
@@ -4,8 +4,6 @@
4 4 5 5
import pandas as pd
6 6 7 -
from .pandas_vb_common import tm
8 - 9 7
for imp in ["pandas.util", "pandas.tools.hashing"]:
10 8
try:
11 9
hashing = import_module(imp)
@@ -47,9 +45,12 @@ def setup(self, unique, sort, dtype):
47 45
elif dtype == "datetime64[ns, tz]":
48 46
data = pd.date_range("2011-01-01", freq="h", periods=N, tz="Asia/Tokyo")
49 47
elif dtype == "object_str":
50 -
data = tm.makeStringIndex(N)
48 +
data = pd.Index([f"i-{i}" for i in range(N)], dtype=object)
51 49
elif dtype == "string[pyarrow]":
52 -
data = pd.array(tm.makeStringIndex(N), dtype="string[pyarrow]")
50 +
data = pd.array(
51 +
pd.Index([f"i-{i}" for i in range(N)], dtype=object),
52 +
dtype="string[pyarrow]",
53 +
)
53 54
else:
54 55
raise NotImplementedError
55 56
@@ -88,7 +89,7 @@ def setup(self, unique, keep, dtype):
88 89
elif dtype == "float64":
89 90
data = pd.Index(np.random.randn(N), dtype="float64")
90 91
elif dtype == "string":
91 -
data = tm.makeStringIndex(N)
92 +
data = pd.Index([f"i-{i}" for i in range(N)], dtype=object)
92 93
elif dtype == "datetime64[ns]":
93 94
data = pd.date_range("2011-01-01", freq="h", periods=N)
94 95
elif dtype == "datetime64[ns, tz]":
@@ -136,7 +137,9 @@ def setup_cache(self):
136 137
df = pd.DataFrame(
137 138
{
138 139
"strings": pd.Series(
139 -
tm.makeStringIndex(10000).take(np.random.randint(0, 10000, size=N))
140 +
pd.Index([f"i-{i}" for i in range(10000)], dtype=object).take(
141 +
np.random.randint(0, 10000, size=N)
142 +
)
140 143
),
141 144
"floats": np.random.randn(N),
142 145
"ints": np.arange(N),
@@ -8,8 +8,6 @@
8 8
date_range,
9 9
)
10 10 11 -
from ..pandas_vb_common import tm
12 - 13 11 14 12
class IsIn:
15 13
params = [
@@ -60,7 +58,9 @@ def setup(self, dtype):
60 58 61 59
elif dtype in ["str", "string[python]", "string[pyarrow]"]:
62 60
try:
63 -
self.series = Series(tm.makeStringIndex(N), dtype=dtype)
61 +
self.series = Series(
62 +
Index([f"i-{i}" for i in range(N)], dtype=object), dtype=dtype
63 +
)
64 64
except ImportError:
65 65
raise NotImplementedError
66 66
self.values = list(self.series[:2])
@@ -9,8 +9,6 @@
9 9
date_range,
10 10
)
11 11 12 -
from .pandas_vb_common import tm
13 - 14 12 15 13
def no_change(arr):
16 14
return arr
@@ -115,7 +113,7 @@ def time_dtindex_from_index_with_series(self):
115 113
class MultiIndexConstructor:
116 114
def setup(self):
117 115
N = 10**4
118 -
self.iterables = [tm.makeStringIndex(N), range(20)]
116 +
self.iterables = [Index([f"i-{i}" for i in range(N)], dtype=object), range(20)]
119 117 120 118
def time_multiindex_from_iterables(self):
121 119
MultiIndex.from_product(self.iterables)
@@ -3,7 +3,10 @@
3 3
import numpy as np
4 4 5 5
import pandas as pd
6 -
from pandas import DataFrame
6 +
from pandas import (
7 +
DataFrame,
8 +
Index,
9 +
)
7 10
import pandas._testing as tm
8 11
from pandas.api.types import (
9 12
is_extension_array_dtype,
@@ -73,8 +76,8 @@ class SelectDtypes:
73 76 74 77
def setup(self, dtype):
75 78
N, K = 5000, 50
76 -
self.index = tm.makeStringIndex(N)
77 -
self.columns = tm.makeStringIndex(K)
79 +
self.index = Index([f"i-{i}" for i in range(N)], dtype=object)
80 +
self.columns = Index([f"i-{i}" for i in range(K)], dtype=object)
78 81 79 82
def create_df(data):
80 83
return DataFrame(data, index=self.index, columns=self.columns)
@@ -12,8 +12,6 @@
12 12
date_range,
13 13
)
14 14 15 -
from .pandas_vb_common import tm
16 - 17 15
try:
18 16
from pandas.tseries.offsets import (
19 17
Hour,
@@ -30,8 +28,8 @@
30 28
class FromDicts:
31 29
def setup(self):
32 30
N, K = 5000, 50
33 -
self.index = tm.makeStringIndex(N)
34 -
self.columns = tm.makeStringIndex(K)
31 +
self.index = pd.Index([f"i-{i}" for i in range(N)], dtype=object)
32 +
self.columns = pd.Index([f"i-{i}" for i in range(K)], dtype=object)
35 33
frame = DataFrame(np.random.randn(N, K), index=self.index, columns=self.columns)
36 34
self.data = frame.to_dict()
37 35
self.dict_list = frame.to_dict(orient="records")
@@ -5,6 +5,7 @@
5 5 6 6
from pandas import (
7 7
DataFrame,
8 +
Index,
8 9
MultiIndex,
9 10
NaT,
10 11
Series,
@@ -14,8 +15,6 @@
14 15
timedelta_range,
15 16
)
16 17 17 -
from .pandas_vb_common import tm
18 - 19 18 20 19
class AsType:
21 20
params = [
@@ -703,8 +702,12 @@ def setup(self, monotonic):
703 702
K = 10
704 703
df = DataFrame(
705 704
{
706 -
"key1": tm.makeStringIndex(N).values.repeat(K),
707 -
"key2": tm.makeStringIndex(N).values.repeat(K),
705 +
"key1": Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(
706 +
K
707 +
),
708 +
"key2": Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(
709 +
K
710 +
),
708 711
"value": np.random.randn(N * K),
709 712
}
710 713
)
@@ -5,15 +5,14 @@
5 5 6 6
from pandas import (
7 7
DataFrame,
8 +
Index,
8 9
Series,
9 10
date_range,
10 11
factorize,
11 12
read_csv,
12 13
)
13 14
from pandas.core.algorithms import take_nd
14 15 15 -
from .pandas_vb_common import tm
16 - 17 16
try:
18 17
from pandas import (
19 18
rolling_kurt,
@@ -34,7 +33,6 @@
34 33
except ImportError:
35 34
from pandas import algos
36 35 37 - 38 36
from .pandas_vb_common import BaseIO # isort:skip
39 37 40 38
@@ -305,7 +303,7 @@ class ParallelFactorize:
305 303
param_names = ["threads"]
306 304 307 305
def setup(self, threads):
308 -
strings = tm.makeStringIndex(100000)
306 +
strings = Index([f"i-{i}" for i in range(100000)], dtype=object)
309 307 310 308
@test_parallel(num_threads=threads)
311 309
def parallel():
@@ -17,8 +17,6 @@
17 17
to_timedelta,
18 18
)
19 19 20 -
from .pandas_vb_common import tm
21 - 22 20
method_blocklist = {
23 21
"object": {
24 22
"diff",
@@ -167,10 +165,14 @@ def setup_cache(self):
167 165
"int64_small": Series(np.random.randint(0, 100, size=size)),
168 166
"int64_large": Series(np.random.randint(0, 10000, size=size)),
169 167
"object_small": Series(
170 -
tm.makeStringIndex(100).take(np.random.randint(0, 100, size=size))
168 +
Index([f"i-{i}" for i in range(100)], dtype=object).take(
169 +
np.random.randint(0, 100, size=size)
170 +
)
171 171
),
172 172
"object_large": Series(
173 -
tm.makeStringIndex(10000).take(np.random.randint(0, 10000, size=size))
173 +
Index([f"i-{i}" for i in range(10000)], dtype=object).take(
174 +
np.random.randint(0, 10000, size=size)
175 +
)
174 176
),
175 177
}
176 178
return data
@@ -912,7 +914,7 @@ def setup(self):
912 914
n1 = 400
913 915
n2 = 250
914 916
index = MultiIndex(
915 -
levels=[np.arange(n1), tm.makeStringIndex(n2)],
917 +
levels=[np.arange(n1), Index([f"i-{i}" for i in range(n2)], dtype=object)],
916 918
codes=[np.repeat(range(n1), n2).tolist(), list(range(n2)) * n1],
917 919
names=["lev1", "lev2"],
918 920
)
@@ -12,8 +12,6 @@
12 12
date_range,
13 13
)
14 14 15 -
from .pandas_vb_common import tm
16 - 17 15 18 16
class SetOperations:
19 17
params = (
@@ -30,7 +28,7 @@ def setup(self, index_structure, dtype, method):
30 28
date_str_left = Index(dates_left.strftime(fmt))
31 29
int_left = Index(np.arange(N))
32 30
ea_int_left = Index(np.arange(N), dtype="Int64")
33 -
str_left = tm.makeStringIndex(N)
31 +
str_left = Index([f"i-{i}" for i in range(N)], dtype=object)
34 32 35 33
data = {
36 34
"datetime": dates_left,
@@ -155,7 +153,12 @@ class Indexing:
155 153 156 154
def setup(self, dtype):
157 155
N = 10**6
158 -
self.idx = getattr(tm, f"make{dtype}Index")(N)
156 +
if dtype == "String":
157 +
self.idx = Index([f"i-{i}" for i in range(N)], dtype=object)
158 +
elif dtype == "Float":
159 +
self.idx = Index(np.arange(N), dtype=np.float64)
160 +
elif dtype == "Int":
161 +
self.idx = Index(np.arange(N), dtype=np.int64)
159 162
self.array_mask = (np.arange(N) % 3) == 0
160 163
self.series_mask = Series(self.array_mask)
161 164
self.sorted = self.idx.sort_values()
@@ -22,8 +22,6 @@
22 22
period_range,
23 23
)
24 24 25 -
from .pandas_vb_common import tm
26 - 27 25 28 26
class NumericSeriesIndexing:
29 27
params = [
@@ -124,7 +122,7 @@ class NonNumericSeriesIndexing:
124 122
def setup(self, index, index_structure):
125 123
N = 10**6
126 124
if index == "string":
127 -
index = tm.makeStringIndex(N)
125 +
index = Index([f"i-{i}" for i in range(N)], dtype=object)
128 126
elif index == "datetime":
129 127
index = date_range("1900", periods=N, freq="s")
130 128
elif index == "period":
@@ -156,8 +154,8 @@ def time_getitem_list_like(self, index, index_structure):
156 154 157 155
class DataFrameStringIndexing:
158 156
def setup(self):
159 -
index = tm.makeStringIndex(1000)
160 -
columns = tm.makeStringIndex(30)
157 +
index = Index([f"i-{i}" for i in range(1000)], dtype=object)
158 +
columns = Index([f"i-{i}" for i in range(30)], dtype=object)
161 159
with warnings.catch_warnings(record=True):
162 160
self.df = DataFrame(np.random.randn(1000, 30), index=index, columns=columns)
163 161
self.idx_scalar = index[100]
You can’t perform that action at this time.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4