/
apply_func.py
293 lines (263 loc) · 11.3 KB
/
apply_func.py
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
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
# Copyright The PyTorch Lightning team.
# Licensed under the Apache License, Version 2.0 (the "License");
# http://www.apache.org/licenses/LICENSE-2.0
#
import dataclasses
from collections import OrderedDict, defaultdict
from copy import deepcopy
from typing import Any, Callable, List, Mapping, Optional, Sequence, Tuple, Union
def is_namedtuple(obj: object) -> bool:
"""Check if object is type nametuple."""
# https://github.com/pytorch/pytorch/blob/v1.8.1/torch/nn/parallel/scatter_gather.py#L4-L8
return isinstance(obj, tuple) and hasattr(obj, "_asdict") and hasattr(obj, "_fields")
def is_dataclass_instance(obj: object) -> bool:
"""Check if object is dataclass."""
# https://docs.python.org/3/library/dataclasses.html#module-level-decorators-classes-and-functions
return dataclasses.is_dataclass(obj) and not isinstance(obj, type)
def apply_to_collection(
data: Any,
dtype: Union[type, Any, Tuple[Union[type, Any]]],
function: Callable,
*args: Any,
wrong_dtype: Optional[Union[type, Tuple[type, ...]]] = None,
include_none: bool = True,
allow_frozen: bool = False,
**kwargs: Any,
) -> Any:
"""Recursively applies a function to all elements of a certain dtype.
Args:
data: the collection to apply the function to
dtype: the given function will be applied to all elements of this dtype
function: the function to apply
*args: positional arguments (will be forwarded to calls of ``function``)
wrong_dtype: the given function won't be applied if this type is specified and the given collections
is of the ``wrong_dtype`` even if it is of type ``dtype``
include_none: Whether to include an element if the output of ``function`` is ``None``.
allow_frozen: Whether not to error upon encountering a frozen dataclass instance.
**kwargs: keyword arguments (will be forwarded to calls of ``function``)
Returns:
The resulting collection
"""
if include_none is False or wrong_dtype is not None or allow_frozen is True:
# not worth implementing these on the fast path: go with the slower option
return _apply_to_collection_slow(
data,
dtype,
function,
*args,
wrong_dtype=wrong_dtype,
include_none=include_none,
allow_frozen=allow_frozen,
**kwargs,
)
# fast path for the most common cases:
if isinstance(data, dtype): # single element
return function(data, *args, **kwargs)
if isinstance(data, list) and all(isinstance(x, dtype) for x in data): # 1d homogeneous list
return [function(x, *args, **kwargs) for x in data]
if isinstance(data, tuple) and all(isinstance(x, dtype) for x in data): # 1d homogeneous tuple
return tuple(function(x, *args, **kwargs) for x in data)
if isinstance(data, dict) and all(isinstance(x, dtype) for x in data.values()): # 1d homogeneous dict
return {k: function(v, *args, **kwargs) for k, v in data.items()}
# slow path for everything else
return _apply_to_collection_slow(
data,
dtype,
function,
*args,
wrong_dtype=wrong_dtype,
include_none=include_none,
allow_frozen=allow_frozen,
**kwargs,
)
def _apply_to_collection_slow(
data: Any,
dtype: Union[type, Any, Tuple[Union[type, Any]]],
function: Callable,
*args: Any,
wrong_dtype: Optional[Union[type, Tuple[type, ...]]] = None,
include_none: bool = True,
allow_frozen: bool = False,
**kwargs: Any,
) -> Any:
# Breaking condition
if isinstance(data, dtype) and (wrong_dtype is None or not isinstance(data, wrong_dtype)):
return function(data, *args, **kwargs)
elem_type = type(data)
# Recursively apply to collection items
if isinstance(data, Mapping):
out = []
for k, v in data.items():
v = _apply_to_collection_slow(
v,
dtype,
function,
*args,
wrong_dtype=wrong_dtype,
include_none=include_none,
allow_frozen=allow_frozen,
**kwargs,
)
if include_none or v is not None:
out.append((k, v))
if isinstance(data, defaultdict):
return elem_type(data.default_factory, OrderedDict(out))
return elem_type(OrderedDict(out))
is_namedtuple_ = is_namedtuple(data)
is_sequence = isinstance(data, Sequence) and not isinstance(data, str)
if is_namedtuple_ or is_sequence:
out = []
for d in data:
v = _apply_to_collection_slow(
d,
dtype,
function,
*args,
wrong_dtype=wrong_dtype,
include_none=include_none,
allow_frozen=allow_frozen,
**kwargs,
)
if include_none or v is not None:
out.append(v)
return elem_type(*out) if is_namedtuple_ else elem_type(out)
if is_dataclass_instance(data):
# make a deepcopy of the data,
# but do not deepcopy mapped fields since the computation would
# be wasted on values that likely get immediately overwritten
fields = {}
memo = {}
for field in dataclasses.fields(data):
field_value = getattr(data, field.name)
fields[field.name] = (field_value, field.init)
memo[id(field_value)] = field_value
result = deepcopy(data, memo=memo)
# apply function to each field
for field_name, (field_value, field_init) in fields.items():
v = None
if field_init:
v = _apply_to_collection_slow(
field_value,
dtype,
function,
*args,
wrong_dtype=wrong_dtype,
include_none=include_none,
allow_frozen=allow_frozen,
**kwargs,
)
if not field_init or (not include_none and v is None): # retain old value
v = getattr(data, field_name)
try:
setattr(result, field_name, v)
except dataclasses.FrozenInstanceError as e:
if allow_frozen:
# Quit early if we encounter a frozen data class; return `result` as is.
break
raise ValueError(
"A frozen dataclass was passed to `apply_to_collection` but this is not allowed."
) from e
return result
# data is neither of dtype, nor a collection
return data
def apply_to_collections(
data1: Optional[Any],
data2: Optional[Any],
dtype: Union[type, Any, Tuple[Union[type, Any]]],
function: Callable,
*args: Any,
wrong_dtype: Optional[Union[type, Tuple[type]]] = None,
**kwargs: Any,
) -> Any:
"""Zips two collections and applies a function to their items of a certain dtype.
Args:
data1: The first collection
data2: The second collection
dtype: the given function will be applied to all elements of this dtype
function: the function to apply
*args: positional arguments (will be forwarded to calls of ``function``)
wrong_dtype: the given function won't be applied if this type is specified and the given collections
is of the ``wrong_dtype`` even if it is of type ``dtype``
**kwargs: keyword arguments (will be forwarded to calls of ``function``)
Returns:
The resulting collection
Raises:
AssertionError:
If sequence collections have different data sizes.
"""
if data1 is None:
if data2 is None:
return None
# in case they were passed reversed
data1, data2 = data2, None
elem_type = type(data1)
if isinstance(data1, dtype) and data2 is not None and (wrong_dtype is None or not isinstance(data1, wrong_dtype)):
return function(data1, data2, *args, **kwargs)
if isinstance(data1, Mapping) and data2 is not None:
# use union because we want to fail if a key does not exist in both
zipped = {k: (data1[k], data2[k]) for k in data1.keys() | data2.keys()}
return elem_type(
{
k: apply_to_collections(*v, dtype, function, *args, wrong_dtype=wrong_dtype, **kwargs)
for k, v in zipped.items()
}
)
is_namedtuple_ = is_namedtuple(data1)
is_sequence = isinstance(data1, Sequence) and not isinstance(data1, str)
if (is_namedtuple_ or is_sequence) and data2 is not None:
if len(data1) != len(data2):
raise ValueError("Sequence collections have different sizes.")
out = [
apply_to_collections(v1, v2, dtype, function, *args, wrong_dtype=wrong_dtype, **kwargs)
for v1, v2 in zip(data1, data2)
]
return elem_type(*out) if is_namedtuple_ else elem_type(out)
if is_dataclass_instance(data1) and data2 is not None:
if not is_dataclass_instance(data2):
raise TypeError(
"Expected inputs to be dataclasses of the same type or to have identical fields"
f" but got input 1 of type {type(data1)} and input 2 of type {type(data2)}."
)
if not (
len(dataclasses.fields(data1)) == len(dataclasses.fields(data2))
and all(map(lambda f1, f2: isinstance(f1, type(f2)), dataclasses.fields(data1), dataclasses.fields(data2)))
):
raise TypeError("Dataclasses fields do not match.")
# make a deepcopy of the data,
# but do not deepcopy mapped fields since the computation would
# be wasted on values that likely get immediately overwritten
data = [data1, data2]
fields: List[dict] = [{}, {}]
memo: dict = {}
for i in range(len(data)):
for field in dataclasses.fields(data[i]):
field_value = getattr(data[i], field.name)
fields[i][field.name] = (field_value, field.init)
if i == 0:
memo[id(field_value)] = field_value
result = deepcopy(data1, memo=memo)
# apply function to each field
for (field_name, (field_value1, field_init1)), (_, (field_value2, field_init2)) in zip(
fields[0].items(), fields[1].items()
):
v = None
if field_init1 and field_init2:
v = apply_to_collections(
field_value1,
field_value2,
dtype,
function,
*args,
wrong_dtype=wrong_dtype,
**kwargs,
)
if not field_init1 or not field_init2 or v is None: # retain old value
return apply_to_collection(data1, dtype, function, *args, wrong_dtype=wrong_dtype, **kwargs)
try:
setattr(result, field_name, v)
except dataclasses.FrozenInstanceError as e:
raise ValueError(
"A frozen dataclass was passed to `apply_to_collections` but this is not allowed."
) from e
return result
return apply_to_collection(data1, dtype, function, *args, wrong_dtype=wrong_dtype, **kwargs)