ProgramComputer
commited on
Commit
•
e1e898e
1
Parent(s):
670c546
Update test.py
Browse files
test.py
CHANGED
@@ -24,11 +24,11 @@ from itertools import repeat
|
|
24 |
from multiprocessing import Manager, Pool, Process
|
25 |
from pathlib import Path
|
26 |
from shutil import copyfileobj
|
27 |
-
|
28 |
import pandas as pd
|
29 |
import requests
|
30 |
from typing import Any, Callable, Dict, Iterable, List, Optional, Set, Tuple, TypeVar, Union
|
31 |
-
|
32 |
import datasets
|
33 |
import urllib3
|
34 |
|
@@ -126,99 +126,6 @@ class NestedDataStructure:
|
|
126 |
else:
|
127 |
return [data]
|
128 |
|
129 |
-
def map_nested(
|
130 |
-
function: Callable[[Any], Any],
|
131 |
-
data_struct: Any,
|
132 |
-
dict_only: bool = False,
|
133 |
-
map_list: bool = True,
|
134 |
-
map_tuple: bool = False,
|
135 |
-
map_numpy: bool = False,
|
136 |
-
num_proc: Optional[int] = None,
|
137 |
-
parallel_min_length: int = 2,
|
138 |
-
types: Optional[tuple] = None,
|
139 |
-
disable_tqdm: bool = True,
|
140 |
-
desc: Optional[str] = None,
|
141 |
-
) -> Any:
|
142 |
-
"""Apply a function recursively to each element of a nested data struct.
|
143 |
-
|
144 |
-
Use multiprocessing if num_proc > 1 and the length of data_struct is greater than or equal to
|
145 |
-
`parallel_min_length`.
|
146 |
-
|
147 |
-
<Changed version="2.5.0">
|
148 |
-
|
149 |
-
Before version 2.5.0, multiprocessing was not used if `num_proc` was greater than or equal to ``len(iterable)``.
|
150 |
-
|
151 |
-
Now, if `num_proc` is greater than or equal to ``len(iterable)``, `num_proc` is set to ``len(iterable)`` and
|
152 |
-
multiprocessing is used.
|
153 |
-
|
154 |
-
</Changed>
|
155 |
-
|
156 |
-
Args:
|
157 |
-
function (`Callable`): Function to be applied to `data_struct`.
|
158 |
-
data_struct (`Any`): Data structure to apply `function` to.
|
159 |
-
dict_only (`bool`, default `False`): Whether only apply `function` recursively to `dict` values in
|
160 |
-
`data_struct`.
|
161 |
-
map_list (`bool`, default `True`): Whether also apply `function` recursively to `list` elements (besides `dict`
|
162 |
-
values).
|
163 |
-
map_tuple (`bool`, default `False`): Whether also apply `function` recursively to `tuple` elements (besides
|
164 |
-
`dict` values).
|
165 |
-
map_numpy (`bool, default `False`): Whether also apply `function` recursively to `numpy.array` elements (besides
|
166 |
-
`dict` values).
|
167 |
-
num_proc (`int`, *optional*): Number of processes.
|
168 |
-
parallel_min_length (`int`, default `2`): Minimum length of `data_struct` required for parallel
|
169 |
-
processing.
|
170 |
-
<Added version="2.5.0"/>
|
171 |
-
types (`tuple`, *optional*): Additional types (besides `dict` values) to apply `function` recursively to their
|
172 |
-
elements.
|
173 |
-
disable_tqdm (`bool`, default `True`): Whether to disable the tqdm progressbar.
|
174 |
-
desc (`str`, *optional*): Prefix for the tqdm progressbar.
|
175 |
-
|
176 |
-
Returns:
|
177 |
-
`Any`
|
178 |
-
"""
|
179 |
-
if types is None:
|
180 |
-
types = []
|
181 |
-
if not dict_only:
|
182 |
-
if map_list:
|
183 |
-
types.append(list)
|
184 |
-
if map_tuple:
|
185 |
-
types.append(tuple)
|
186 |
-
if map_numpy:
|
187 |
-
types.append(np.ndarray)
|
188 |
-
types = tuple(types)
|
189 |
-
|
190 |
-
# Singleton
|
191 |
-
if not isinstance(data_struct, dict) and not isinstance(data_struct, types):
|
192 |
-
return function(data_struct)
|
193 |
-
|
194 |
-
disable_tqdm = disable_tqdm or not logging.is_progress_bar_enabled()
|
195 |
-
iterable = list(data_struct.values()) if isinstance(data_struct, dict) else data_struct
|
196 |
-
|
197 |
-
if num_proc is None:
|
198 |
-
num_proc = 1
|
199 |
-
if num_proc != -1 and num_proc <= 1 or len(iterable) < parallel_min_length:
|
200 |
-
mapped = [
|
201 |
-
_single_map_nested((function, obj, types, None, True, None))
|
202 |
-
for obj in logging.tqdm(iterable, disable=disable_tqdm, desc=desc)
|
203 |
-
]
|
204 |
-
else:
|
205 |
-
with warnings.catch_warnings():
|
206 |
-
warnings.filterwarnings(
|
207 |
-
"ignore",
|
208 |
-
message=".* is experimental and might be subject to breaking changes in the future\\.$",
|
209 |
-
category=UserWarning,
|
210 |
-
)
|
211 |
-
mapped = parallel_map(function, iterable, num_proc, types, disable_tqdm, desc, _single_map_nested)
|
212 |
-
|
213 |
-
if isinstance(data_struct, dict):
|
214 |
-
return dict(zip(data_struct.keys(), mapped))
|
215 |
-
else:
|
216 |
-
if isinstance(data_struct, list):
|
217 |
-
return mapped
|
218 |
-
elif isinstance(data_struct, tuple):
|
219 |
-
return tuple(mapped)
|
220 |
-
else:
|
221 |
-
return np.array(mapped)
|
222 |
|
223 |
def _mp_download(
|
224 |
url,
|
|
|
24 |
from multiprocessing import Manager, Pool, Process
|
25 |
from pathlib import Path
|
26 |
from shutil import copyfileobj
|
27 |
+
import numpy as np
|
28 |
import pandas as pd
|
29 |
import requests
|
30 |
from typing import Any, Callable, Dict, Iterable, List, Optional, Set, Tuple, TypeVar, Union
|
31 |
+
import warnings
|
32 |
import datasets
|
33 |
import urllib3
|
34 |
|
|
|
126 |
else:
|
127 |
return [data]
|
128 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
129 |
|
130 |
def _mp_download(
|
131 |
url,
|