forked from pytorch/pytorch
/
module.py
2047 lines (1661 loc) · 85.1 KB
/
module.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
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
from collections import OrderedDict, namedtuple
import itertools
import warnings
import functools
import weakref
import torch
from ..parameter import Parameter
import torch.utils.hooks as hooks
from torch import Tensor, device, dtype
from typing import Union, Tuple, Any, Callable, Iterator, Set, Optional, overload, TypeVar, Mapping, Dict, List
from ...utils.hooks import RemovableHandle
__all__ = ['register_module_forward_pre_hook', 'register_module_forward_hook', 'register_module_backward_hook',
'register_module_full_backward_hook', 'Module']
_grad_t = Union[Tuple[Tensor, ...], Tensor]
# See https://mypy.readthedocs.io/en/latest/generics.html#generic-methods-and-generic-self for the use
# of `T` to annotate `self`. Many methods of `Module` return `self` and we want those return values to be
# the type of the subclass, not the looser type of `Module`.
T = TypeVar('T', bound='Module')
class _IncompatibleKeys(namedtuple('IncompatibleKeys', ['missing_keys', 'unexpected_keys'])):
def __repr__(self):
if not self.missing_keys and not self.unexpected_keys:
return '<All keys matched successfully>'
return super(_IncompatibleKeys, self).__repr__()
__str__ = __repr__
def _addindent(s_, numSpaces):
s = s_.split('\n')
# don't do anything for single-line stuff
if len(s) == 1:
return s_
first = s.pop(0)
s = [(numSpaces * ' ') + line for line in s]
s = '\n'.join(s)
s = first + '\n' + s
return s
class _WrappedHook:
def __init__(self, hook: Callable, module: Optional["Module"] = None):
self.hook: Callable = hook
functools.update_wrapper(self, hook)
self.with_module: bool = False
if module is not None:
self.module: weakref.ReferenceType["Module"] = weakref.ref(module)
self.with_module = True
def __call__(self, *args: Any, **kwargs: Any) -> Any:
if self.with_module:
module = self.module()
if module is None:
raise RuntimeError("You are trying to call the hook of a dead Module!")
return self.hook(module, *args, **kwargs)
return self.hook(*args, **kwargs)
def __getstate__(self) -> Dict:
result = {"hook": self.hook, "with_module": self.with_module}
if self.with_module:
result["module"] = self.module()
return result
def __setstate__(self, state: Dict):
self.hook = state["hook"]
self.with_module = state["with_module"]
if self.with_module:
if state["module"] is None:
raise RuntimeError("You are trying to revive the hook of a dead Module!")
self.module = weakref.ref(state["module"])
r"""This tracks hooks common to all modules that are executed before/after
calling forward and backward. This is global state used for debugging/profiling
purposes"""
_global_backward_hooks: Dict[int, Callable] = OrderedDict()
_global_is_full_backward_hook: Optional[bool] = None
_global_forward_pre_hooks: Dict[int, Callable] = OrderedDict()
_global_forward_hooks: Dict[int, Callable] = OrderedDict()
_EXTRA_STATE_KEY_SUFFIX = '_extra_state'
def register_module_forward_pre_hook(hook: Callable[..., None]) -> RemovableHandle:
r"""Registers a forward pre-hook common to all modules.
.. warning ::
This adds global state to the `nn.module` module
and it is only intended for debugging/profiling purposes.
The hook will be called every time before :func:`forward` is invoked.
It should have the following signature::
hook(module, input) -> None or modified input
The input contains only the positional arguments given to the module.
Keyword arguments won't be passed to the hooks and only to the ``forward``.
The hook can modify the input. User can either return a tuple or a
single modified value in the hook. We will wrap the value into a tuple
if a single value is returned(unless that value is already a tuple).
This hook has precedence over the specific module hooks registered with
``register_forward_pre_hook``.
Returns:
:class:`torch.utils.hooks.RemovableHandle`:
a handle that can be used to remove the added hook by calling
``handle.remove()``
"""
handle = hooks.RemovableHandle(_global_forward_pre_hooks)
_global_forward_pre_hooks[handle.id] = hook
return handle
def register_module_forward_hook(hook: Callable[..., None]) -> RemovableHandle:
r"""Registers a global forward hook for all the modules
.. warning ::
This adds global state to the `nn.module` module
and it is only intended for debugging/profiling purposes.
The hook will be called every time after :func:`forward` has computed an output.
It should have the following signature::
hook(module, input, output) -> None or modified output
The input contains only the positional arguments given to the module.
Keyword arguments won't be passed to the hooks and only to the ``forward``.
The hook can modify the output. It can modify the input inplace but
it will not have effect on forward since this is called after
:func:`forward` is called.
Returns:
:class:`torch.utils.hooks.RemovableHandle`:
a handle that can be used to remove the added hook by calling
``handle.remove()``
This hook will be executed before specific module hooks registered with
``register_forward_hook``.
"""
handle = hooks.RemovableHandle(_global_forward_hooks)
_global_forward_hooks[handle.id] = hook
return handle
def register_module_backward_hook(
hook: Callable[['Module', _grad_t, _grad_t], Union[None, Tensor]]
) -> RemovableHandle:
r"""Registers a backward hook common to all the modules.
This function is deprecated in favor of
:func:`torch.nn.modules.module.register_module_full_backward_hook`
and the behavior of this function will change in future versions.
Returns:
:class:`torch.utils.hooks.RemovableHandle`:
a handle that can be used to remove the added hook by calling
``handle.remove()``
"""
global _global_is_full_backward_hook
if _global_is_full_backward_hook is True:
raise RuntimeError("Cannot use both regular backward hooks and full backward hooks as a "
"global Module hook. Please use only one of them.")
_global_is_full_backward_hook = False
handle = hooks.RemovableHandle(_global_backward_hooks)
_global_backward_hooks[handle.id] = hook
return handle
def register_module_full_backward_hook(
hook: Callable[['Module', _grad_t, _grad_t], Union[None, Tensor]]
) -> RemovableHandle:
r"""Registers a backward hook common to all the modules.
.. warning ::
This adds global state to the `nn.module` module
and it is only intended for debugging/profiling purposes.
The hook will be called every time the gradients with respect to a module
are computed, i.e. the hook will execute if and only if the gradients with
respect to module outputs are computed. The hook should have the following
signature::
hook(module, grad_input, grad_output) -> Tensor or None
The :attr:`grad_input` and :attr:`grad_output` are tuples. The hook should
not modify its arguments, but it can optionally return a new gradient with
respect to the input that will be used in place of :attr:`grad_input` in
subsequent computations. :attr:`grad_input` will only correspond to the inputs given
as positional arguments and all kwarg arguments will not appear in the hook. Entries
in :attr:`grad_input` and :attr:`grad_output` will be ``None`` for all non-Tensor
arguments.
For technical reasons, when this hook is applied to a Module, its forward function will
receive a view of each Tensor passed to the Module. Similarly the caller will receive a view
of each Tensor returned by the Module's forward function.
Global hooks are called before hooks registered with `register_backward_hook`
Returns:
:class:`torch.utils.hooks.RemovableHandle`:
a handle that can be used to remove the added hook by calling
``handle.remove()``
"""
global _global_is_full_backward_hook
if _global_is_full_backward_hook is False:
raise RuntimeError("Cannot use both regular backward hooks and full backward hooks as a "
"global Module hook. Please use only one of them.")
_global_is_full_backward_hook = True
handle = hooks.RemovableHandle(_global_backward_hooks)
_global_backward_hooks[handle.id] = hook
return handle
# Trick mypy into not applying contravariance rules to inputs by defining
# forward as a value, rather than a function. See also
# https://github.com/python/mypy/issues/8795
def _forward_unimplemented(self, *input: Any) -> None:
r"""Defines the computation performed at every call.
Should be overridden by all subclasses.
.. note::
Although the recipe for forward pass needs to be defined within
this function, one should call the :class:`Module` instance afterwards
instead of this since the former takes care of running the
registered hooks while the latter silently ignores them.
"""
raise NotImplementedError(f"Module [{type(self).__name__}] is missing the required \"forward\" function")
class Module:
r"""Base class for all neural network modules.
Your models should also subclass this class.
Modules can also contain other Modules, allowing to nest them in
a tree structure. You can assign the submodules as regular attributes::
import torch.nn as nn
import torch.nn.functional as F
class Model(nn.Module):
def __init__(self):
super().__init__()
self.conv1 = nn.Conv2d(1, 20, 5)
self.conv2 = nn.Conv2d(20, 20, 5)
def forward(self, x):
x = F.relu(self.conv1(x))
return F.relu(self.conv2(x))
Submodules assigned in this way will be registered, and will have their
parameters converted too when you call :meth:`to`, etc.
.. note::
As per the example above, an ``__init__()`` call to the parent class
must be made before assignment on the child.
:ivar training: Boolean represents whether this module is in training or
evaluation mode.
:vartype training: bool
"""
dump_patches: bool = False
_version: int = 1
r"""This allows better BC support for :meth:`load_state_dict`. In
:meth:`state_dict`, the version number will be saved as in the attribute
`_metadata` of the returned state dict, and thus pickled. `_metadata` is a
dictionary with keys that follow the naming convention of state dict. See
``_load_from_state_dict`` on how to use this information in loading.
If new parameters/buffers are added/removed from a module, this number shall
be bumped, and the module's `_load_from_state_dict` method can compare the
version number and do appropriate changes if the state dict is from before
the change."""
training: bool
_parameters: Dict[str, Optional[Parameter]]
_buffers: Dict[str, Optional[Tensor]]
_non_persistent_buffers_set: Set[str]
_backward_hooks: Dict[int, Callable]
_is_full_backward_hook: Optional[bool]
_forward_hooks: Dict[int, Callable]
_forward_pre_hooks: Dict[int, Callable]
_state_dict_hooks: Dict[int, Callable]
_load_state_dict_pre_hooks: Dict[int, Callable]
_load_state_dict_post_hooks: Dict[int, Callable]
_modules: Dict[str, Optional['Module']]
def __init__(self) -> None:
"""
Initializes internal Module state, shared by both nn.Module and ScriptModule.
"""
torch._C._log_api_usage_once("python.nn_module")
"""
Calls super().__setattr__('a', a) instead of the typical self.a = a
to avoid Module.__setattr__ overhead. Module's __setattr__ has special
handling for parameters, submodules, and buffers but simply calls into
super().__setattr__ for all other attributes.
"""
super().__setattr__('training', True)
super().__setattr__('_parameters', OrderedDict())
super().__setattr__('_buffers', OrderedDict())
super().__setattr__('_non_persistent_buffers_set', set())
super().__setattr__('_backward_hooks', OrderedDict())
super().__setattr__('_is_full_backward_hook', None)
super().__setattr__('_forward_hooks', OrderedDict())
super().__setattr__('_forward_pre_hooks', OrderedDict())
super().__setattr__('_state_dict_hooks', OrderedDict())
super().__setattr__('_load_state_dict_pre_hooks', OrderedDict())
super().__setattr__('_load_state_dict_post_hooks', OrderedDict())
super().__setattr__('_modules', OrderedDict())
forward: Callable[..., Any] = _forward_unimplemented
def register_buffer(self, name: str, tensor: Optional[Tensor], persistent: bool = True) -> None:
r"""Adds a buffer to the module.
This is typically used to register a buffer that should not to be
considered a model parameter. For example, BatchNorm's ``running_mean``
is not a parameter, but is part of the module's state. Buffers, by
default, are persistent and will be saved alongside parameters. This
behavior can be changed by setting :attr:`persistent` to ``False``. The
only difference between a persistent buffer and a non-persistent buffer
is that the latter will not be a part of this module's
:attr:`state_dict`.
Buffers can be accessed as attributes using given names.
Args:
name (str): name of the buffer. The buffer can be accessed
from this module using the given name
tensor (Tensor or None): buffer to be registered. If ``None``, then operations
that run on buffers, such as :attr:`cuda`, are ignored. If ``None``,
the buffer is **not** included in the module's :attr:`state_dict`.
persistent (bool): whether the buffer is part of this module's
:attr:`state_dict`.
Example::
>>> # xdoctest: +SKIP("undefined vars")
>>> self.register_buffer('running_mean', torch.zeros(num_features))
"""
if persistent is False and isinstance(self, torch.jit.ScriptModule):
raise RuntimeError("ScriptModule does not support non-persistent buffers")
if '_buffers' not in self.__dict__:
raise AttributeError(
"cannot assign buffer before Module.__init__() call")
elif not isinstance(name, torch._six.string_classes):
raise TypeError("buffer name should be a string. "
"Got {}".format(torch.typename(name)))
elif '.' in name:
raise KeyError("buffer name can't contain \".\"")
elif name == '':
raise KeyError("buffer name can't be empty string \"\"")
elif hasattr(self, name) and name not in self._buffers:
raise KeyError("attribute '{}' already exists".format(name))
elif tensor is not None and not isinstance(tensor, torch.Tensor):
raise TypeError("cannot assign '{}' object to buffer '{}' "
"(torch Tensor or None required)"
.format(torch.typename(tensor), name))
else:
self._buffers[name] = tensor
if persistent:
self._non_persistent_buffers_set.discard(name)
else:
self._non_persistent_buffers_set.add(name)
def register_parameter(self, name: str, param: Optional[Parameter]) -> None:
r"""Adds a parameter to the module.
The parameter can be accessed as an attribute using given name.
Args:
name (str): name of the parameter. The parameter can be accessed
from this module using the given name
param (Parameter or None): parameter to be added to the module. If
``None``, then operations that run on parameters, such as :attr:`cuda`,
are ignored. If ``None``, the parameter is **not** included in the
module's :attr:`state_dict`.
"""
if '_parameters' not in self.__dict__:
raise AttributeError(
"cannot assign parameter before Module.__init__() call")
elif not isinstance(name, torch._six.string_classes):
raise TypeError("parameter name should be a string. "
"Got {}".format(torch.typename(name)))
elif '.' in name:
raise KeyError("parameter name can't contain \".\"")
elif name == '':
raise KeyError("parameter name can't be empty string \"\"")
elif hasattr(self, name) and name not in self._parameters:
raise KeyError("attribute '{}' already exists".format(name))
if param is None:
self._parameters[name] = None
elif not isinstance(param, Parameter):
raise TypeError("cannot assign '{}' object to parameter '{}' "
"(torch.nn.Parameter or None required)"
.format(torch.typename(param), name))
elif param.grad_fn:
raise ValueError(
"Cannot assign non-leaf Tensor to parameter '{0}'. Model "
"parameters must be created explicitly. To express '{0}' "
"as a function of another Tensor, compute the value in "
"the forward() method.".format(name))
else:
self._parameters[name] = param
def add_module(self, name: str, module: Optional['Module']) -> None:
r"""Adds a child module to the current module.
The module can be accessed as an attribute using the given name.
Args:
name (str): name of the child module. The child module can be
accessed from this module using the given name
module (Module): child module to be added to the module.
"""
if not isinstance(module, Module) and module is not None:
raise TypeError("{} is not a Module subclass".format(
torch.typename(module)))
elif not isinstance(name, torch._six.string_classes):
raise TypeError("module name should be a string. Got {}".format(
torch.typename(name)))
elif hasattr(self, name) and name not in self._modules:
raise KeyError("attribute '{}' already exists".format(name))
elif '.' in name:
raise KeyError("module name can't contain \".\", got: {}".format(name))
elif name == '':
raise KeyError("module name can't be empty string \"\"")
self._modules[name] = module
def register_module(self, name: str, module: Optional['Module']) -> None:
r"""Alias for :func:`add_module`."""
self.add_module(name, module)
def get_submodule(self, target: str) -> "Module":
"""
Returns the submodule given by ``target`` if it exists,
otherwise throws an error.
For example, let's say you have an ``nn.Module`` ``A`` that
looks like this:
.. code-block:: text
A(
(net_b): Module(
(net_c): Module(
(conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))
)
(linear): Linear(in_features=100, out_features=200, bias=True)
)
)
(The diagram shows an ``nn.Module`` ``A``. ``A`` has a nested
submodule ``net_b``, which itself has two submodules ``net_c``
and ``linear``. ``net_c`` then has a submodule ``conv``.)
To check whether or not we have the ``linear`` submodule, we
would call ``get_submodule("net_b.linear")``. To check whether
we have the ``conv`` submodule, we would call
``get_submodule("net_b.net_c.conv")``.
The runtime of ``get_submodule`` is bounded by the degree
of module nesting in ``target``. A query against
``named_modules`` achieves the same result, but it is O(N) in
the number of transitive modules. So, for a simple check to see
if some submodule exists, ``get_submodule`` should always be
used.
Args:
target: The fully-qualified string name of the submodule
to look for. (See above example for how to specify a
fully-qualified string.)
Returns:
torch.nn.Module: The submodule referenced by ``target``
Raises:
AttributeError: If the target string references an invalid
path or resolves to something that is not an
``nn.Module``
"""
if target == "":
return self
atoms: List[str] = target.split(".")
mod: torch.nn.Module = self
for item in atoms:
if not hasattr(mod, item):
raise AttributeError(mod._get_name() + " has no "
"attribute `" + item + "`")
mod = getattr(mod, item)
if not isinstance(mod, torch.nn.Module):
raise AttributeError("`" + item + "` is not "
"an nn.Module")
return mod
def get_parameter(self, target: str) -> "Parameter":
"""
Returns the parameter given by ``target`` if it exists,
otherwise throws an error.
See the docstring for ``get_submodule`` for a more detailed
explanation of this method's functionality as well as how to
correctly specify ``target``.
Args:
target: The fully-qualified string name of the Parameter
to look for. (See ``get_submodule`` for how to specify a
fully-qualified string.)
Returns:
torch.nn.Parameter: The Parameter referenced by ``target``
Raises:
AttributeError: If the target string references an invalid
path or resolves to something that is not an
``nn.Parameter``
"""
module_path, _, param_name = target.rpartition(".")
mod: torch.nn.Module = self.get_submodule(module_path)
if not hasattr(mod, param_name):
raise AttributeError(mod._get_name() + " has no attribute `"
+ param_name + "`")
param: torch.nn.Parameter = getattr(mod, param_name)
if not isinstance(param, torch.nn.Parameter):
raise AttributeError("`" + param_name + "` is not an "
"nn.Parameter")
return param
def get_buffer(self, target: str) -> "Tensor":
"""
Returns the buffer given by ``target`` if it exists,
otherwise throws an error.
See the docstring for ``get_submodule`` for a more detailed
explanation of this method's functionality as well as how to
correctly specify ``target``.
Args:
target: The fully-qualified string name of the buffer
to look for. (See ``get_submodule`` for how to specify a
fully-qualified string.)
Returns:
torch.Tensor: The buffer referenced by ``target``
Raises:
AttributeError: If the target string references an invalid
path or resolves to something that is not a
buffer
"""
module_path, _, buffer_name = target.rpartition(".")
mod: torch.nn.Module = self.get_submodule(module_path)
if not hasattr(mod, buffer_name):
raise AttributeError(mod._get_name() + " has no attribute `"
+ buffer_name + "`")
buffer: torch.Tensor = getattr(mod, buffer_name)
if buffer_name not in mod._buffers:
raise AttributeError("`" + buffer_name + "` is not a buffer")
return buffer
def get_extra_state(self) -> Any:
"""
Returns any extra state to include in the module's state_dict.
Implement this and a corresponding :func:`set_extra_state` for your module
if you need to store extra state. This function is called when building the
module's `state_dict()`.
Note that extra state should be pickleable to ensure working serialization
of the state_dict. We only provide provide backwards compatibility guarantees
for serializing Tensors; other objects may break backwards compatibility if
their serialized pickled form changes.
Returns:
object: Any extra state to store in the module's state_dict
"""
raise RuntimeError(
"Reached a code path in Module.get_extra_state() that should never be called. "
"Please file an issue at https://github.com/pytorch/pytorch/issues/new?template=bug-report.yml "
"to report this bug.")
def set_extra_state(self, state: Any):
"""
This function is called from :func:`load_state_dict` to handle any extra state
found within the `state_dict`. Implement this function and a corresponding
:func:`get_extra_state` for your module if you need to store extra state within its
`state_dict`.
Args:
state (dict): Extra state from the `state_dict`
"""
raise RuntimeError(
"Reached a code path in Module.set_extra_state() that should never be called. "
"Please file an issue at https://github.com/pytorch/pytorch/issues/new?template=bug-report.yml "
"to report this bug.")
def _apply(self, fn):
for module in self.children():
module._apply(fn)
def compute_should_use_set_data(tensor, tensor_applied):
if torch._has_compatible_shallow_copy_type(tensor, tensor_applied):
# If the new tensor has compatible tensor type as the existing tensor,
# the current behavior is to change the tensor in-place using `.data =`,
# and the future behavior is to overwrite the existing tensor. However,
# changing the current behavior is a BC-breaking change, and we want it
# to happen in future releases. So for now we introduce the
# `torch.__future__.get_overwrite_module_params_on_conversion()`
# global flag to let the user control whether they want the future
# behavior of overwriting the existing tensor or not.
return not torch.__future__.get_overwrite_module_params_on_conversion()
else:
return False
for key, param in self._parameters.items():
if param is None:
continue
# Tensors stored in modules are graph leaves, and we don't want to
# track autograd history of `param_applied`, so we have to use
# `with torch.no_grad():`
with torch.no_grad():
param_applied = fn(param)
should_use_set_data = compute_should_use_set_data(param, param_applied)
if should_use_set_data:
param.data = param_applied
out_param = param
else:
assert isinstance(param, Parameter)
assert param.is_leaf
out_param = Parameter(param_applied, param.requires_grad)
self._parameters[key] = out_param
if param.grad is not None:
with torch.no_grad():
grad_applied = fn(param.grad)
should_use_set_data = compute_should_use_set_data(param.grad, grad_applied)
if should_use_set_data:
assert out_param.grad is not None
out_param.grad.data = grad_applied
else:
assert param.grad.is_leaf
out_param.grad = grad_applied.requires_grad_(param.grad.requires_grad)
for key, buf in self._buffers.items():
if buf is not None:
self._buffers[key] = fn(buf)
return self
def apply(self: T, fn: Callable[['Module'], None]) -> T:
r"""Applies ``fn`` recursively to every submodule (as returned by ``.children()``)
as well as self. Typical use includes initializing the parameters of a model
(see also :ref:`nn-init-doc`).
Args:
fn (:class:`Module` -> None): function to be applied to each submodule
Returns:
Module: self
Example::
>>> @torch.no_grad()
>>> def init_weights(m):
>>> print(m)
>>> if type(m) == nn.Linear:
>>> m.weight.fill_(1.0)
>>> print(m.weight)
>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
[1., 1.]], requires_grad=True)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[1., 1.],
[1., 1.]], requires_grad=True)
Sequential(
(0): Linear(in_features=2, out_features=2, bias=True)
(1): Linear(in_features=2, out_features=2, bias=True)
)
"""
for module in self.children():
module.apply(fn)
fn(self)
return self
def cuda(self: T, device: Optional[Union[int, device]] = None) -> T:
r"""Moves all model parameters and buffers to the GPU.
This also makes associated parameters and buffers different objects. So
it should be called before constructing optimizer if the module will
live on GPU while being optimized.
.. note::
This method modifies the module in-place.
Args:
device (int, optional): if specified, all parameters will be
copied to that device
Returns:
Module: self
"""
return self._apply(lambda t: t.cuda(device))
def ipu(self: T, device: Optional[Union[int, device]] = None) -> T:
r"""Moves all model parameters and buffers to the IPU.
This also makes associated parameters and buffers different objects. So
it should be called before constructing optimizer if the module will
live on IPU while being optimized.
.. note::
This method modifies the module in-place.
Arguments:
device (int, optional): if specified, all parameters will be
copied to that device
Returns:
Module: self
"""
return self._apply(lambda t: t.ipu(device))
def xpu(self: T, device: Optional[Union[int, device]] = None) -> T:
r"""Moves all model parameters and buffers to the XPU.
This also makes associated parameters and buffers different objects. So
it should be called before constructing optimizer if the module will
live on XPU while being optimized.
.. note::
This method modifies the module in-place.
Arguments:
device (int, optional): if specified, all parameters will be
copied to that device
Returns:
Module: self
"""
return self._apply(lambda t: t.xpu(device))
def cpu(self: T) -> T:
r"""Moves all model parameters and buffers to the CPU.
.. note::
This method modifies the module in-place.
Returns:
Module: self
"""
return self._apply(lambda t: t.cpu())
def type(self: T, dst_type: Union[dtype, str]) -> T:
r"""Casts all parameters and buffers to :attr:`dst_type`.
.. note::
This method modifies the module in-place.
Args:
dst_type (type or string): the desired type
Returns:
Module: self
"""
return self._apply(lambda t: t.type(dst_type))
def float(self: T) -> T:
r"""Casts all floating point parameters and buffers to ``float`` datatype.
.. note::
This method modifies the module in-place.
Returns:
Module: self
"""
return self._apply(lambda t: t.float() if t.is_floating_point() else t)
def double(self: T) -> T:
r"""Casts all floating point parameters and buffers to ``double`` datatype.
.. note::
This method modifies the module in-place.
Returns:
Module: self
"""
return self._apply(lambda t: t.double() if t.is_floating_point() else t)
def half(self: T) -> T:
r"""Casts all floating point parameters and buffers to ``half`` datatype.
.. note::
This method modifies the module in-place.
Returns:
Module: self
"""
return self._apply(lambda t: t.half() if t.is_floating_point() else t)
def bfloat16(self: T) -> T:
r"""Casts all floating point parameters and buffers to ``bfloat16`` datatype.
.. note::
This method modifies the module in-place.
Returns:
Module: self
"""
return self._apply(lambda t: t.bfloat16() if t.is_floating_point() else t)
def to_empty(self: T, *, device: Union[str, device]) -> T:
r"""Moves the parameters and buffers to the specified device without copying storage.
Args:
device (:class:`torch.device`): The desired device of the parameters
and buffers in this module.
Returns:
Module: self
"""
return self._apply(lambda t: torch.empty_like(t, device=device))
@overload
def to(self: T, device: Optional[Union[int, device]] = ..., dtype: Optional[Union[dtype, str]] = ...,
non_blocking: bool = ...) -> T:
...
@overload
def to(self: T, dtype: Union[dtype, str], non_blocking: bool = ...) -> T:
...
@overload
def to(self: T, tensor: Tensor, non_blocking: bool = ...) -> T:
...
def to(self, *args, **kwargs):
r"""Moves and/or casts the parameters and buffers.
This can be called as
.. function:: to(device=None, dtype=None, non_blocking=False)
:noindex:
.. function:: to(dtype, non_blocking=False)
:noindex:
.. function:: to(tensor, non_blocking=False)
:noindex:
.. function:: to(memory_format=torch.channels_last)
:noindex:
Its signature is similar to :meth:`torch.Tensor.to`, but only accepts
floating point or complex :attr:`dtype`\ s. In addition, this method will
only cast the floating point or complex parameters and buffers to :attr:`dtype`
(if given). The integral parameters and buffers will be moved
:attr:`device`, if that is given, but with dtypes unchanged. When
:attr:`non_blocking` is set, it tries to convert/move asynchronously
with respect to the host if possible, e.g., moving CPU Tensors with
pinned memory to CUDA devices.
See below for examples.
.. note::
This method modifies the module in-place.
Args:
device (:class:`torch.device`): the desired device of the parameters
and buffers in this module
dtype (:class:`torch.dtype`): the desired floating point or complex dtype of
the parameters and buffers in this module
tensor (torch.Tensor): Tensor whose dtype and device are the desired
dtype and device for all parameters and buffers in this module
memory_format (:class:`torch.memory_format`): the desired memory
format for 4D parameters and buffers in this module (keyword
only argument)
Returns:
Module: self
Examples::
>>> # xdoctest: +IGNORE_WANT("non-deterministic")
>>> linear = nn.Linear(2, 2)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
[-0.5113, -0.2325]])
>>> linear.to(torch.double)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
[-0.5113, -0.2325]], dtype=torch.float64)
>>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA1)
>>> gpu1 = torch.device("cuda:1")
>>> linear.to(gpu1, dtype=torch.half, non_blocking=True)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
[-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1')
>>> cpu = torch.device("cpu")
>>> linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
[-0.5112, -0.2324]], dtype=torch.float16)
>>> linear = nn.Linear(2, 2, bias=None).to(torch.cdouble)
>>> linear.weight
Parameter containing:
tensor([[ 0.3741+0.j, 0.2382+0.j],
[ 0.5593+0.j, -0.4443+0.j]], dtype=torch.complex128)
>>> linear(torch.ones(3, 2, dtype=torch.cdouble))
tensor([[0.6122+0.j, 0.1150+0.j],
[0.6122+0.j, 0.1150+0.j],
[0.6122+0.j, 0.1150+0.j]], dtype=torch.complex128)
"""
device, dtype, non_blocking, convert_to_format = torch._C._nn._parse_to(*args, **kwargs)
if dtype is not None:
if not (dtype.is_floating_point or dtype.is_complex):
raise TypeError('nn.Module.to only accepts floating point or complex '
'dtypes, but got desired dtype={}'.format(dtype))
if dtype.is_complex:
warnings.warn(
"Complex modules are a new feature under active development whose design may change, "
"and some modules might not work as expected when using complex tensors as parameters or buffers. "
"Please file an issue at https://github.com/pytorch/pytorch/issues/new?template=bug-report.yml "
"if a complex module does not work as expected.")
def convert(t):
if convert_to_format is not None and t.dim() in (4, 5):
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None,
non_blocking, memory_format=convert_to_format)
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
return self._apply(convert)
def register_backward_hook(
self, hook: Callable[['Module', _grad_t, _grad_t], Union[None, Tensor]]
) -> RemovableHandle:
r"""Registers a backward hook on the module.
This function is deprecated in favor of :meth:`~torch.nn.Module.register_full_backward_hook` and
the behavior of this function will change in future versions.
Returns:
:class:`torch.utils.hooks.RemovableHandle`: