1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
|
import numpy as np
from pandas import (
DataFrame,
Index,
MultiIndex,
Series,
date_range,
period_range,
)
class Reindex:
def setup(self):
rng = date_range(start="1/1/1970", periods=10000, freq="1min")
self.df = DataFrame(np.random.rand(10000, 10), index=rng, columns=range(10))
self.df["foo"] = "bar"
self.rng_subset = Index(rng[::2])
self.df2 = DataFrame(
index=range(10000), data=np.random.rand(10000, 30), columns=range(30)
)
N = 5000
K = 200
level1 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
level2 = np.tile(Index([f"i-{i}" for i in range(K)], dtype=object).values, N)
index = MultiIndex.from_arrays([level1, level2])
self.s = Series(np.random.randn(N * K), index=index)
self.s_subset = self.s[::2]
self.s_subset_no_cache = self.s[::2].copy()
mi = MultiIndex.from_product([rng, range(100)])
self.s2 = Series(np.random.randn(len(mi)), index=mi)
self.s2_subset = self.s2[::2].copy()
def time_reindex_dates(self):
self.df.reindex(self.rng_subset)
def time_reindex_columns(self):
self.df2.reindex(columns=self.df.columns[1:5])
def time_reindex_multiindex_with_cache(self):
# MultiIndex._values gets cached
self.s.reindex(self.s_subset.index)
def time_reindex_multiindex_no_cache(self):
# Copy to avoid MultiIndex._values getting cached
self.s.reindex(self.s_subset_no_cache.index.copy())
def time_reindex_multiindex_no_cache_dates(self):
# Copy to avoid MultiIndex._values getting cached
self.s2_subset.reindex(self.s2.index.copy())
class ReindexMethod:
params = [["pad", "backfill"], [date_range, period_range]]
param_names = ["method", "constructor"]
def setup(self, method, constructor):
N = 100000
self.idx = constructor("1/1/2000", periods=N, freq="1min")
self.ts = Series(np.random.randn(N), index=self.idx)[::2]
def time_reindex_method(self, method, constructor):
self.ts.reindex(self.idx, method=method)
class LevelAlign:
def setup(self):
self.index = MultiIndex(
levels=[np.arange(10), np.arange(100), np.arange(100)],
codes=[
np.arange(10).repeat(10000),
np.tile(np.arange(100).repeat(100), 10),
np.tile(np.tile(np.arange(100), 100), 10),
],
)
self.df = DataFrame(np.random.randn(len(self.index), 4), index=self.index)
self.df_level = DataFrame(np.random.randn(100, 4), index=self.index.levels[1])
def time_align_level(self):
self.df.align(self.df_level, level=1, copy=False)
def time_reindex_level(self):
self.df_level.reindex(self.index, level=1)
class DropDuplicates:
params = [True, False]
param_names = ["inplace"]
def setup(self, inplace):
N = 10000
K = 10
key1 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
key2 = Index([f"i-{i}" for i in range(N)], dtype=object).values.repeat(K)
self.df = DataFrame(
{"key1": key1, "key2": key2, "value": np.random.randn(N * K)}
)
self.df_nan = self.df.copy()
self.df_nan.iloc[:10000, :] = np.nan
self.s = Series(np.random.randint(0, 1000, size=10000))
self.s_str = Series(
np.tile(Index([f"i-{i}" for i in range(1000)], dtype=object).values, 10)
)
N = 1000000
K = 10000
key1 = np.random.randint(0, K, size=N)
self.df_int = DataFrame({"key1": key1})
self.df_bool = DataFrame(np.random.randint(0, 2, size=(K, 10), dtype=bool))
def time_frame_drop_dups(self, inplace):
self.df.drop_duplicates(["key1", "key2"], inplace=inplace)
def time_frame_drop_dups_na(self, inplace):
self.df_nan.drop_duplicates(["key1", "key2"], inplace=inplace)
def time_series_drop_dups_int(self, inplace):
self.s.drop_duplicates(inplace=inplace)
def time_series_drop_dups_string(self, inplace):
self.s_str.drop_duplicates(inplace=inplace)
def time_frame_drop_dups_int(self, inplace):
self.df_int.drop_duplicates(inplace=inplace)
def time_frame_drop_dups_bool(self, inplace):
self.df_bool.drop_duplicates(inplace=inplace)
class Align:
# blog "pandas escaped the zoo"
def setup(self):
n = 50000
indices = Index([f"i-{i}" for i in range(n)], dtype=object)
subsample_size = 40000
self.x = Series(np.random.randn(n), indices)
self.y = Series(
np.random.randn(subsample_size),
index=np.random.choice(indices, subsample_size, replace=False),
)
def time_align_series_irregular_string(self):
self.x + self.y
from .pandas_vb_common import setup # noqa: F401 isort:skip
|
