File size: 72,403 Bytes
9c6594c |
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 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 |
# mypy: allow-untyped-defs
import copyreg
import enum
import functools
import warnings
from collections import OrderedDict
from copy import deepcopy
from numbers import Number
from typing import Any, Callable, cast, Optional, Union
import torch
import torch._C as _C
from torch._namedtensor_internals import (
check_serializing_named_tensor,
is_ellipsis,
resolve_ellipsis,
single_ellipsis_index,
unzip_namedshape,
update_names,
)
from torch.overrides import (
get_default_nowrap_functions,
handle_torch_function,
has_torch_function,
has_torch_function_unary,
has_torch_function_variadic,
)
def _handle_torch_function_and_wrap_type_error_to_not_implemented(f):
assigned = functools.WRAPPER_ASSIGNMENTS
@functools.wraps(f, assigned=assigned)
def wrapped(*args, **kwargs):
try:
# See https://github.com/pytorch/pytorch/issues/75462
if has_torch_function(args):
return handle_torch_function(wrapped, args, *args, **kwargs)
return f(*args, **kwargs)
except TypeError:
return NotImplemented
return wrapped
# Should not be used, this is kept only for BC of loading old serialized Tensor subclasses
def _rebuild_from_type(func, type, args, dict):
if type is Tensor:
return func(*args)
ret = func(*args).as_subclass(type)
ret.__dict__ = dict
return ret
def _rebuild_from_type_v2(func, new_type, args, state):
ret = func(*args)
if type(ret) is not new_type:
ret = ret.as_subclass(new_type)
# Tensor does define __setstate__ even though it doesn't define
# __getstate__. So only use __setstate__ if it is NOT the one defined
# on Tensor
if (
getattr(ret.__class__, "__setstate__", Tensor.__setstate__)
is not Tensor.__setstate__
):
ret.__setstate__(state)
else:
ret = torch._utils._set_obj_state(ret, state)
return ret
def _dtype_to_typestr(dtype):
# CUDA devices are little-endian and tensors are stored in native byte
# order. 1-byte entries are endian-agnostic.
return {
torch.complex64: "<c8",
torch.complex128: "<c16",
torch.bfloat16: "<V2", # Same as ml_dtypes.bfloat16.dtype.str.
torch.float16: "<f2",
torch.float32: "<f4",
torch.float64: "<f8",
torch.uint8: "|u1",
torch.int8: "|i1",
torch.uint16: "<u2",
torch.int16: "<i2",
torch.uint32: "<u4",
torch.int32: "<i4",
torch.uint64: "<u8",
torch.int64: "<i8",
torch.bool: "|b1",
}[dtype]
# NB: If you subclass Tensor, and want to share the subclassed class
# across processes, you must also update torch/multiprocessing/reductions.py
# to define a ForkingPickler serialization mode for the class.
#
# NB: If you add a new method to Tensor, you must update
# torch/_C/__init__.pyi.in to add a type annotation for your method;
# otherwise, it will not show up in autocomplete.
class Tensor(torch._C.TensorBase):
_is_param: bool
def _clear_non_serializable_cached_data(self):
r"""Clears any data cached in the tensor's ``__dict__`` that would prevent the tensor
from being serialized.
For example, subclasses with custom dispatched sizes / strides cache this info in
non-serializable PyCapsules within the ``__dict__``, and this must be cleared out for
serialization to function.
Any subclass that overrides this MUST call ``super()._clear_non_serializable_cached_data().``
Additional data cleared within the override must be able to be re-cached transparently
to avoid breaking subclass functionality.
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor._clear_non_serializable_cached_data, (self,), self
)
# NB: Wrapper subclasses that implement custom-dispatched sizes / strides cache
# this info via non-serializable PyCapsules.
CACHED_SIZES_STRIDES_KEYS = [
"_sym_sizes_capsule",
"_sym_sizes_capsule_len",
"_sym_strides_capsule",
"_sym_strides_capsule_len",
]
for key in CACHED_SIZES_STRIDES_KEYS:
self.__dict__.pop(key, None)
def __deepcopy__(self, memo):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__deepcopy__, (self,), self, memo)
if not self.is_leaf:
raise RuntimeError(
"Only Tensors created explicitly by the user "
"(graph leaves) support the deepcopy protocol at the moment. "
"If you were attempting to deepcopy a module, this may be because "
"of a torch.nn.utils.weight_norm usage, "
"see https://github.com/pytorch/pytorch/pull/103001"
)
if id(self) in memo:
return memo[id(self)]
with torch.no_grad():
# TODO: skipping storage copy is wrong for meta, as meta
# does accurate alias tracking; however, the code below
# doesn't work because of
# https://github.com/pytorch/pytorch/issues/47442
# Update the test in test_serialization if you remove 'meta' from here
if (
self.is_sparse
or self.device.type
in ["lazy", "xla", "mtia", "mps", "maia", "meta", "ipu"]
or (
not torch._C._has_storage(self)
and self.device.type == torch._C._get_privateuse1_backend_name()
)
or (type(self) is not Tensor and self.data_ptr() == 0)
):
new_tensor = self.clone()
if type(new_tensor) is not type(self):
raise RuntimeError(
"The default implementation of __deepcopy__() for wrapper subclasses "
"only works for subclass types that implement clone() and for which "
"cloning returns another instance of the same subclass. You should either "
"properly implement clone() for your subclass or override __deepcopy__() "
"if it is intended behavior for clone() to return an instance of a "
"different type."
)
else:
new_storage = self._typed_storage()._deepcopy(memo)
if self.is_quantized:
# quantizer_params can be different type based on torch attribute
quantizer_params: Union[
tuple[torch.qscheme, float, int],
tuple[torch.qscheme, Tensor, Tensor, int],
]
if self.qscheme() == torch.per_tensor_affine:
quantizer_params = (
self.qscheme(),
self.q_scale(),
self.q_zero_point(),
)
elif self.qscheme() in (
torch.per_channel_affine,
torch.per_channel_affine_float_qparams,
):
quantizer_params = (
self.qscheme(),
self.q_per_channel_scales(),
self.q_per_channel_zero_points(),
self.q_per_channel_axis(),
)
else:
raise RuntimeError(
f"Unsupported qscheme {self.qscheme()} in deepcopy"
)
# TODO: Once we decide to break serialization FC, no longer
# need to wrap with TypedStorage
new_tensor = torch._utils._rebuild_qtensor(
torch.storage.TypedStorage(
wrap_storage=new_storage._untyped_storage,
dtype=self.dtype,
_internal=True,
),
self.storage_offset(),
self.size(),
self.stride(),
quantizer_params,
self.requires_grad,
self._backward_hooks,
)
if type(new_tensor) is not type(self):
raise RuntimeError(
"The default implementation of __deepcopy__() for quantized tensors "
"expects the tensor returned by torch._utils._rebuild_qtensor() to "
"match the type of the instance being copied. If you encounter this, "
"please open an issue on PyTorch's GitHub."
)
else:
new_tensor = self.new_empty([])
if type(new_tensor) is not type(self):
raise RuntimeError(
"The default implementation of __deepcopy__() for non-wrapper subclasses "
"only works for subclass types that implement new_empty() and for which "
"that function returns another instance of the same subclass. You should "
"either properly implement new_empty() for your subclass or override "
"__deepcopy__() if it is intended behavior for new_empty() to return "
"an instance of a different type."
)
new_tensor.set_(
new_storage, self.storage_offset(), self.size(), self.stride()
)
if self.is_conj():
new_tensor = new_tensor.conj_physical()
if self.is_neg():
new_tensor = new_tensor.neg()
if self.requires_grad:
new_tensor.requires_grad_()
if self.grad is not None:
new_tensor.grad = self.grad.__deepcopy__(memo)
if type(self) is not Tensor:
if type(new_tensor) is not type(self):
raise RuntimeError(
"Type of deepcopy result does not match the type of the source tensor. "
"If you encounter this, please open an issue on PyTorch's GitHub."
)
# Plain Tensors don't have slots
slots_to_save = copyreg._slotnames(self.__class__) # type: ignore[attr-defined]
for slot in slots_to_save:
if hasattr(self, slot):
setattr(new_tensor, slot, deepcopy(getattr(self, slot), memo))
# don't try to deepcopy non-serializable cached data
self._clear_non_serializable_cached_data()
new_tensor.__dict__ = deepcopy(self.__dict__, memo)
memo[id(self)] = new_tensor
return new_tensor
def __reduce_ex__(self, proto):
materialize_fake_tensors = (
torch.serialization._serialization_tls.materialize_fake_tensors
)
state = torch._utils._get_obj_state(self)
# Ignore all state when using FakeTensor with skip_data(materialize_fake_tensors) because FakeTensor has
# some state that cannot be pickled
if (
# TODO: remove hasattr, it's a hack to support versions of torch that
# don't have _subclasses
hasattr(torch, "_subclasses")
and type(self) is torch._subclasses.fake_tensor.FakeTensor
and materialize_fake_tensors
) or (type(self) is Tensor and not state):
# Fast path for regular tensor without Python state.
return self._reduce_ex_internal(proto)
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__reduce_ex__, (self,), self, proto)
func, args = self._reduce_ex_internal(proto)
# sizes / strides cache needs to be cleared here because it'll just be re-cached
# if cleared earlier. Note that state references the -actual- tensor dict.
self._clear_non_serializable_cached_data()
return (_rebuild_from_type_v2, (func, type(self), args, state))
def storage(self):
r"""
storage() -> torch.TypedStorage
Returns the underlying :class:`TypedStorage`.
.. warning::
:class:`TypedStorage` is deprecated. It will be removed in the future, and
:class:`UntypedStorage` will be the only storage class. To access the
:class:`UntypedStorage` directly, use :attr:`Tensor.untyped_storage()`.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.storage, (self,), self)
torch.storage._warn_typed_storage_removal(stacklevel=2)
return self._typed_storage()
# For internal use only, to avoid raising deprecation warning
def _typed_storage(self):
untyped_storage = self.untyped_storage()
return torch.TypedStorage(
wrap_storage=untyped_storage, dtype=self.dtype, _internal=True
)
def _reduce_ex_internal(self, proto):
check_serializing_named_tensor(self)
from torch.utils.hooks import warn_if_has_hooks
# See Note [Don't serialize hooks]
warn_if_has_hooks(self)
backward_hooks: dict[Any, Any] = OrderedDict()
skip_data = torch.serialization._serialization_tls.skip_data
materialize_fake_tensors = (
torch.serialization._serialization_tls.materialize_fake_tensors
)
if self.device.type in ["xla", "maia"] or (
not torch._C._has_storage(self)
and self.device.type == torch._C._get_privateuse1_backend_name()
):
if skip_data:
raise RuntimeError(
"Cannot serialize tensors on backends with no storage under skip_data context manager"
)
cpu_tensor = self.cpu()
return (
torch._utils._rebuild_device_tensor_from_cpu_tensor,
(cpu_tensor, self.dtype, str(self.device), self.requires_grad),
)
# Legacy comment that does not hold anymore.
# Note: Numpy array is chosen to be the rebuild component for XLA, MTIA, MAIA Tensors.
# We considered a few options:
# 1. CPU tensor can't be used here.
# Otherwise in torch.load CPU storage is reconstructed with randomly
# initialized data, moved onto backend device, and then storage is updated
# to the serialized content. This works perfectly for CPU/CUDA but not these backends;
# their tensors are disconnected with storage so they don't get the update.
# 2. Python list is not a good fit due to performance reason.
# `tolist()` converts every single element in the tensor into python objects
# and serialize them one by one.
if self.device.type in ["mtia"]:
# Convert BFloat16 tesors to Float32 before conversion to numpy, as numpy doesn't
# support BFloat16. The rebuild tensor from numpy takes in the original self.dtype,
# this would reconstruct the BFloat16 tensor from numpy.
if skip_data:
raise RuntimeError(
"Cannot serialize tensors on backends with no storage under skip_data context manager"
)
numpy_tensor = (
self.cpu().numpy()
if self.dtype != torch.bfloat16
else self.cpu().to(torch.float32).numpy()
)
return (
torch._utils._rebuild_device_tensor_from_numpy,
(numpy_tensor, self.dtype, str(self.device), self.requires_grad),
)
if self.device.type == "meta":
# NB: This implementation BREAKS storage sharing. Current
# hypothesis is that no one cares for meta tensors.
if skip_data:
warnings.warn(
"Serializing tensors on the meta device under skip_data context manager is a no-op"
)
arg_meta = (
self.dtype,
tuple(self.size()),
self.stride(),
self.requires_grad,
)
return (torch._utils._rebuild_meta_tensor_no_storage, arg_meta)
if self.is_quantized:
if skip_data:
raise RuntimeError(
"Cannot serialize qtensor under skip_data context manager, file an issue if you need this feature"
)
# quantizer_params can be different type based on torch attribute
quantizer_params: Union[
tuple[torch.qscheme, float, int], tuple[Any, Tensor, Tensor, int]
]
if self.qscheme() == torch.per_tensor_affine:
quantizer_params = (
torch.per_tensor_affine,
self.q_scale(),
self.q_zero_point(),
)
elif self.qscheme() in (
torch.per_channel_affine,
torch.per_channel_affine_float_qparams,
):
# convert scales and zero points to tuple to avoid recursive calls
# when/if we get multi-axis quantized tensors in the future, the shape
# is recoverable from the main tensor shape
quantizer_params = (
torch.per_channel_affine,
self.q_per_channel_scales(),
self.q_per_channel_zero_points(),
self.q_per_channel_axis(),
)
else:
raise RuntimeError(
f"Serialization is not supported for tensors of type {self.qscheme()}"
)
# TODO: Once we decide to break serialization FC, no longer
# need to wrap with TypedStorage
args_qtensor = (
torch.storage.TypedStorage(
wrap_storage=self._typed_storage()._untyped_storage,
dtype=self.dtype,
_internal=True,
),
self.storage_offset(),
tuple(self.size()),
self.stride(),
quantizer_params,
self.requires_grad,
backward_hooks,
)
return (torch._utils._rebuild_qtensor, args_qtensor)
elif self.is_sparse:
if self.layout == torch.sparse_coo:
args_sparse = (
self.layout,
(self._indices(), self._values(), self.size(), self.is_coalesced()),
)
else:
raise NotImplementedError(
f"sparse tensor __reduce_ex__ for layout `{self.layout}`"
)
return (torch._utils._rebuild_sparse_tensor, args_sparse)
elif self.layout in {
torch.sparse_csr,
torch.sparse_csc,
torch.sparse_bsr,
torch.sparse_bsc,
}:
if self.layout in {torch.sparse_csr, torch.sparse_bsr}:
compressed_indices, plain_indices = (
self.crow_indices(),
self.col_indices(),
)
else:
compressed_indices, plain_indices = (
self.ccol_indices(),
self.row_indices(),
)
args_sparse_compressed = (
self.layout,
(
compressed_indices,
plain_indices,
self.values(),
self.size(),
),
)
return (torch._utils._rebuild_sparse_tensor, args_sparse_compressed)
elif self.is_nested:
if skip_data:
raise RuntimeError(
"Cannot serialize nested tensor under skip_data context manager, file an issue if you need this feature"
)
args_nested = (
# NB: values() currently returns the storage as a buffer in an unsafe way.
# Ideally, we'd use a private API for this instead. TODO: Switch to this if
# we ever get around to adding it.
self.values(),
self._nested_tensor_size(),
self._nested_tensor_strides(),
self._nested_tensor_storage_offsets(),
)
return (torch._utils._rebuild_nested_tensor, args_nested)
elif (
type(self) is not torch.Tensor
and type(self).__torch_dispatch__ is not torch.Tensor.__torch_dispatch__
and (
isinstance(self, torch._subclasses.functional_tensor.FunctionalTensor)
or (
not isinstance(self, torch._subclasses.fake_tensor.FakeTensor)
and self.data_ptr() == 0
)
)
):
arg_wrapper_subclass = (
type(self),
self.dtype,
tuple(self.size()),
self.stride(),
self.storage_offset(),
self.layout,
self.device,
self.requires_grad,
)
return (torch._utils._rebuild_wrapper_subclass, arg_wrapper_subclass)
elif (
type(self) is not torch.Tensor
and type(self).__torch_dispatch__ is not torch.Tensor.__torch_dispatch__
and (
isinstance(self, torch._subclasses.fake_tensor.FakeTensor)
and not (skip_data and materialize_fake_tensors)
)
):
arg_wrapper_subclass = (
type(self),
self.dtype,
tuple(self.size()),
self.stride(),
self.storage_offset(),
self.layout,
self.device,
self.requires_grad,
)
return (torch._utils._rebuild_wrapper_subclass, arg_wrapper_subclass)
else:
v3_dtypes = torch.storage._new_dtypes()
if self.dtype in v3_dtypes:
rebuild_func = torch._utils._rebuild_tensor_v3
storage = self.untyped_storage()
else:
# TODO: Once we decide to break serialization FC, no longer
# need to wrap with TypedStorage
rebuild_func = torch._utils._rebuild_tensor_v2 # type: ignore[assignment]
storage = torch.storage.TypedStorage(
wrap_storage=self._typed_storage()._untyped_storage,
dtype=self.dtype,
_internal=True,
) # type: ignore[assignment]
# TODO: remove hasattr, it's a hack to support versions of torch that
# don't have _subclasses
if (
hasattr(torch, "_subclasses")
and isinstance(self, torch._subclasses.fake_tensor.FakeTensor)
and skip_data
):
storage._fake_device = self.device
args = (
storage,
self.storage_offset(),
tuple(self.size()),
self.stride(),
self.requires_grad,
backward_hooks,
) # previously was self._backward_hooks
if isinstance(storage, torch.storage.UntypedStorage):
args = args + (self.dtype,) # type: ignore[assignment]
metadata = torch._utils.get_tensor_metadata(self)
if metadata:
args = args + (metadata,) # type: ignore[assignment]
return (rebuild_func, args)
def __setstate__(self, state):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__setstate__, (self,), self, state)
# Warning: this method is NOT called when you torch.load() a tensor;
# that is managed by _rebuild_tensor_v2
if not self.is_leaf:
raise RuntimeError("__setstate__ can be only called on leaf Tensors")
if len(state) == 4:
# legacy serialization of Tensor
self.set_(*state)
return
elif len(state) == 5:
# legacy serialization of Variable
self.data = state[0]
state = (state[3], state[4], state[2])
# The setting of _backward_hooks is expected to be a no-op.
# See Note [Don't serialize hooks]
self.requires_grad, _, self._backward_hooks = state
def __repr__(self, *, tensor_contents=None):
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.__repr__, (self,), self, tensor_contents=tensor_contents
)
# All strings are unicode in Python 3.
return torch._tensor_str._str(self, tensor_contents=tensor_contents)
def backward(
self, gradient=None, retain_graph=None, create_graph=False, inputs=None
):
r"""Computes the gradient of current tensor wrt graph leaves.
The graph is differentiated using the chain rule. If the tensor is
non-scalar (i.e. its data has more than one element) and requires
gradient, the function additionally requires specifying a ``gradient``.
It should be a tensor of matching type and shape, that represents
the gradient of the differentiated function w.r.t. ``self``.
This function accumulates gradients in the leaves - you might need to zero
``.grad`` attributes or set them to ``None`` before calling it.
See :ref:`Default gradient layouts<default-grad-layouts>`
for details on the memory layout of accumulated gradients.
.. note::
If you run any forward ops, create ``gradient``, and/or call ``backward``
in a user-specified CUDA stream context, see
:ref:`Stream semantics of backward passes<bwd-cuda-stream-semantics>`.
.. note::
When ``inputs`` are provided and a given input is not a leaf,
the current implementation will call its grad_fn (though it is not strictly needed to get this gradients).
It is an implementation detail on which the user should not rely.
See https://github.com/pytorch/pytorch/pull/60521#issuecomment-867061780 for more details.
Args:
gradient (Tensor, optional): The gradient of the function
being differentiated w.r.t. ``self``.
This argument can be omitted if ``self`` is a scalar.
retain_graph (bool, optional): If ``False``, the graph used to compute
the grads will be freed. Note that in nearly all cases setting
this option to True is not needed and often can be worked around
in a much more efficient way. Defaults to the value of
``create_graph``.
create_graph (bool, optional): If ``True``, graph of the derivative will
be constructed, allowing to compute higher order derivative
products. Defaults to ``False``.
inputs (sequence of Tensor, optional): Inputs w.r.t. which the gradient will be
accumulated into ``.grad``. All other tensors will be ignored. If not
provided, the gradient is accumulated into all the leaf Tensors that were
used to compute the :attr:`tensors`.
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.backward,
(self,),
self,
gradient=gradient,
retain_graph=retain_graph,
create_graph=create_graph,
inputs=inputs,
)
torch.autograd.backward(
self, gradient, retain_graph, create_graph, inputs=inputs
)
def register_hook(self, hook):
r"""Registers a backward hook.
The hook will be called every time a gradient with respect to the
Tensor is computed. The hook should have the following signature::
hook(grad) -> Tensor or None
The hook should not modify its argument, but it can optionally return
a new gradient which will be used in place of :attr:`grad`.
This function returns a handle with a method ``handle.remove()``
that removes the hook from the module.
.. note::
See :ref:`backward-hooks-execution` for more information on how when this hook
is executed, and how its execution is ordered relative to other hooks.
Example::
>>> v = torch.tensor([0., 0., 0.], requires_grad=True)
>>> h = v.register_hook(lambda grad: grad * 2) # double the gradient
>>> v.backward(torch.tensor([1., 2., 3.]))
>>> v.grad
2
4
6
[torch.FloatTensor of size (3,)]
>>> h.remove() # removes the hook
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.register_hook, (self,), self, hook)
if not self.requires_grad:
raise RuntimeError(
"cannot register a hook on a tensor that doesn't require gradient"
)
if self._backward_hooks is None:
self._backward_hooks = OrderedDict()
if self.grad_fn is not None:
self.grad_fn._register_hook_dict(self)
from torch.utils.hooks import RemovableHandle
handle = RemovableHandle(self._backward_hooks)
self._backward_hooks[handle.id] = hook
return handle
def register_post_accumulate_grad_hook(self, hook):
r"""Registers a backward hook that runs after grad accumulation.
The hook will be called after all gradients for a tensor have been accumulated,
meaning that the .grad field has been updated on that tensor. The post
accumulate grad hook is ONLY applicable for leaf tensors (tensors without a
.grad_fn field). Registering this hook on a non-leaf tensor will error!
The hook should have the following signature::
hook(param: Tensor) -> None
Note that, unlike other autograd hooks, this hook operates on the tensor
that requires grad and not the grad itself. The hook can in-place modify
and access its Tensor argument, including its .grad field.
This function returns a handle with a method ``handle.remove()``
that removes the hook from the module.
.. note::
See :ref:`backward-hooks-execution` for more information on how when this hook
is executed, and how its execution is ordered relative to other hooks. Since
this hook runs during the backward pass, it will run in no_grad mode (unless
create_graph is True). You can use torch.enable_grad() to re-enable autograd
within the hook if you need it.
Example::
>>> v = torch.tensor([0., 0., 0.], requires_grad=True)
>>> lr = 0.01
>>> # simulate a simple SGD update
>>> h = v.register_post_accumulate_grad_hook(lambda p: p.add_(p.grad, alpha=-lr))
>>> v.backward(torch.tensor([1., 2., 3.]))
>>> v
tensor([-0.0100, -0.0200, -0.0300], requires_grad=True)
>>> h.remove() # removes the hook
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.register_post_accumulate_grad_hook, (self,), self, hook
)
if not self.requires_grad:
raise RuntimeError(
"cannot register a hook on a tensor that doesn't require gradient"
)
if self.grad_fn is not None:
raise RuntimeError(
"post accumulate grad hooks cannot be registered on non-leaf tensors"
)
if self._post_accumulate_grad_hooks is None:
self._post_accumulate_grad_hooks: dict[Any, Any] = OrderedDict()
from torch.utils.hooks import RemovableHandle
handle = RemovableHandle(self._post_accumulate_grad_hooks)
self._post_accumulate_grad_hooks[handle.id] = hook
return handle
def reinforce(self, reward):
def trim(str):
return "\n".join([line.strip() for line in str.split("\n")])
raise RuntimeError(
trim(
r"""reinforce() was removed.
Use torch.distributions instead.
See https://pytorch.org/docs/main/distributions.html
Instead of:
probs = policy_network(state)
action = probs.multinomial()
next_state, reward = env.step(action)
action.reinforce(reward)
action.backward()
Use:
probs = policy_network(state)
# NOTE: categorical is equivalent to what used to be called multinomial
m = torch.distributions.Categorical(probs)
action = m.sample()
next_state, reward = env.step(action)
loss = -m.log_prob(action) * reward
loss.backward()
"""
)
)
detach = _C._add_docstr(
_C.TensorBase.detach,
r"""
Returns a new Tensor, detached from the current graph.
The result will never require gradient.
This method also affects forward mode AD gradients and the result will never
have forward mode AD gradients.
.. note::
Returned Tensor shares the same storage with the original one.
In-place modifications on either of them will be seen, and may trigger
errors in correctness checks.
""",
)
detach_ = _C._add_docstr(
_C.TensorBase.detach_,
r"""
Detaches the Tensor from the graph that created it, making it a leaf.
Views cannot be detached in-place.
This method also affects forward mode AD gradients and the result will never
have forward mode AD gradients.
""",
)
def is_shared(self):
r"""Checks if tensor is in shared memory.
This is always ``True`` for CUDA tensors.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.is_shared, (self,), self)
return self._typed_storage()._is_shared()
def share_memory_(self):
r"""Moves the underlying storage to shared memory.
This is a no-op if the underlying storage is already in shared memory
and for CUDA tensors. Tensors in shared memory cannot be resized.
See :meth:`torch.UntypedStorage.share_memory_` for more details.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.share_memory_, (self,), self)
self._typed_storage()._share_memory_()
return self
def module_load(self, other, assign=False):
r"""Defines how to transform ``other`` when loading it into ``self`` in :meth:`~nn.Module.load_state_dict`.
Used when :func:`~torch.__future__.get_swap_module_params_on_conversion` is ``True``.
It is expected that ``self`` is a parameter or buffer in an ``nn.Module`` and ``other`` is the
value in the state dictionary with the corresponding key, this method defines
how ``other`` is remapped before being swapped with ``self`` via
:func:`~torch.utils.swap_tensors` in :meth:`~nn.Module.load_state_dict`.
.. note::
This method should always return a new object that is not ``self`` or ``other``.
For example, the default implementation returns ``self.copy_(other).detach()``
if ``assign`` is ``False`` or ``other.detach()`` if ``assign`` is ``True``.
Args:
other (Tensor): value in state dict with key corresponding to ``self``
assign (bool): the assign argument passed to :meth:`nn.Module.load_state_dict`
"""
if has_torch_function_variadic(self, other):
return handle_torch_function(
Tensor.module_load, (self, other), self, other, assign=assign
)
if assign:
return other.detach()
else:
return self.copy_(other).detach()
def __reversed__(self):
r"""Reverses the tensor along dimension 0."""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__reversed__, (self,), self)
if self.dim() == 0:
return self
else:
return self.flip(0)
def norm(
self,
p: Optional[Union[float, str]] = "fro",
dim=None,
keepdim=False,
dtype=None,
):
r"""See :func:`torch.norm`"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.norm, (self,), self, p=p, dim=dim, keepdim=keepdim, dtype=dtype
)
return torch.norm(self, p, dim, keepdim, dtype=dtype)
def solve(self, other):
from torch._linalg_utils import solve
return solve(self, other)
def lstsq(self, other):
from torch._linalg_utils import lstsq
return lstsq(self, other)
def eig(self, eigenvectors=False):
from torch._linalg_utils import eig
return eig(self, eigenvectors=eigenvectors)
def symeig(self, eigenvectors=False):
from torch._linalg_utils import _symeig
return _symeig(self, eigenvectors=eigenvectors)
def lu(self, pivot=True, get_infos=False):
r"""See :func:`torch.lu`"""
# If get_infos is True, then we don't need to check for errors and vice versa
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.lu, (self,), self, pivot=pivot, get_infos=get_infos
)
LU, pivots, infos = torch._lu_with_info(
self, pivot=pivot, check_errors=(not get_infos)
)
if get_infos:
return LU, pivots, infos
else:
return LU, pivots
def stft(
self,
n_fft: int,
hop_length: Optional[int] = None,
win_length: Optional[int] = None,
window: "Optional[Tensor]" = None,
center: bool = True,
pad_mode: str = "reflect",
normalized: bool = False,
onesided: Optional[bool] = None,
return_complex: Optional[bool] = None,
align_to_window: Optional[bool] = None,
):
r"""See :func:`torch.stft`
.. warning::
This function changed signature at version 0.4.1. Calling with
the previous signature may cause error or return incorrect result.
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.stft,
(self,),
self,
n_fft,
hop_length=hop_length,
win_length=win_length,
window=window,
center=center,
pad_mode=pad_mode,
normalized=normalized,
onesided=onesided,
return_complex=return_complex,
align_to_window=align_to_window,
)
return torch.stft(
self,
n_fft,
hop_length,
win_length,
window,
center,
pad_mode,
normalized,
onesided,
return_complex=return_complex,
align_to_window=align_to_window,
)
def istft(
self,
n_fft: int,
hop_length: Optional[int] = None,
win_length: Optional[int] = None,
window: "Optional[Tensor]" = None,
center: bool = True,
normalized: bool = False,
onesided: Optional[bool] = None,
length: Optional[int] = None,
return_complex: bool = False,
):
r"""See :func:`torch.istft`"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.istft,
(self,),
self,
n_fft,
hop_length=hop_length,
win_length=win_length,
window=window,
center=center,
normalized=normalized,
onesided=onesided,
length=length,
return_complex=return_complex,
)
return torch.istft(
self,
n_fft,
hop_length,
win_length,
window,
center,
normalized,
onesided,
length,
return_complex=return_complex,
)
def resize(self, *sizes):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.resize, (self,), self, *sizes)
warnings.warn("non-inplace resize is deprecated")
from torch.autograd._functions import Resize
return Resize.apply(self, sizes)
def resize_as(self, tensor):
if has_torch_function_variadic(self, tensor):
return handle_torch_function(Tensor.resize_as, (self, tensor), self, tensor)
warnings.warn("non-inplace resize_as is deprecated")
from torch.autograd._functions import Resize
return Resize.apply(self, tensor.size())
def split(self, split_size, dim=0):
r"""See :func:`torch.split`"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.split, (self,), self, split_size, dim=dim
)
if isinstance(split_size, Tensor):
try:
split_size = int(split_size)
except ValueError:
pass
if isinstance(split_size, (int, torch.SymInt)):
return torch._VF.split(self, split_size, dim) # type: ignore[attr-defined]
else:
return torch._VF.split_with_sizes(self, split_size, dim)
def unique(self, sorted=True, return_inverse=False, return_counts=False, dim=None):
r"""Returns the unique elements of the input tensor.
See :func:`torch.unique`
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.unique,
(self,),
self,
sorted=sorted,
return_inverse=return_inverse,
return_counts=return_counts,
dim=dim,
)
return torch.unique(
self,
sorted=sorted,
return_inverse=return_inverse,
return_counts=return_counts,
dim=dim,
)
def unique_consecutive(self, return_inverse=False, return_counts=False, dim=None):
r"""Eliminates all but the first element from every consecutive group of equivalent elements.
See :func:`torch.unique_consecutive`
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.unique_consecutive,
(self,),
self,
return_inverse=return_inverse,
return_counts=return_counts,
dim=dim,
)
return torch.unique_consecutive(
self, return_inverse=return_inverse, return_counts=return_counts, dim=dim
)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rsub__(self, other):
return _C._VariableFunctions.rsub(self, other)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rdiv__(self, other):
return self.reciprocal() * other
__rtruediv__ = __rdiv__
__itruediv__ = _C.TensorBase.__idiv__
__pow__ = cast(
Callable[
["torch._C.TensorBase", Union["Tensor", int, float, bool, complex]],
"Tensor",
],
_handle_torch_function_and_wrap_type_error_to_not_implemented(
_C.TensorBase.pow
),
)
__ipow__ = _handle_torch_function_and_wrap_type_error_to_not_implemented(
_C.TensorBase.pow_
)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rmod__(self, other):
return torch.remainder(other, self)
def __format__(self, format_spec):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__format__, (self,), self, format_spec)
if self.dim() == 0 and not self.is_meta and type(self) is Tensor:
return self.item().__format__(format_spec)
return object.__format__(self, format_spec)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rpow__(self, other):
return torch.pow(other, self)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __floordiv__(self, other):
return torch.floor_divide(self, other)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rfloordiv__(self, other):
return torch.floor_divide(other, self)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rlshift__(self, other):
return torch.bitwise_left_shift(other, self)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rrshift__(self, other):
return torch.bitwise_right_shift(other, self)
@_handle_torch_function_and_wrap_type_error_to_not_implemented
def __rmatmul__(self, other):
return torch.matmul(other, self)
__pos__ = _C.TensorBase.positive
__neg__ = _C.TensorBase.neg
__abs__ = _C.TensorBase.abs
def __len__(self):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__len__, (self,), self)
if self.dim() == 0:
raise TypeError("len() of a 0-d tensor")
if torch._C._get_tracing_state():
warnings.warn(
"Using len to get tensor shape might cause the trace to be incorrect. "
"Recommended usage would be tensor.shape[0]. "
"Passing a tensor of different shape might lead to errors or silently give "
"incorrect results.",
category=torch.jit.TracerWarning,
stacklevel=2,
)
return self.shape[0]
def __iter__(self):
# NB: we use 'imap' and not 'map' here, so that in Python 2 we get a
# generator and don't eagerly perform all the indexes. This could
# save us work, and also helps keep trace ordering deterministic
# (e.g., if you zip(*hiddens), the eager map will force all the
# indexes of hiddens[0] before hiddens[1], while the generator
# map will interleave them.)
# NB: We have intentionally skipped __torch_function__ dispatch here.
# See gh-54457
if self.dim() == 0:
raise TypeError("iteration over a 0-d tensor")
if torch._C._get_tracing_state():
warnings.warn(
"Iterating over a tensor might cause the trace to be incorrect. "
"Passing a tensor of different shape won't change the number of "
"iterations executed (and might lead to errors or silently give "
"incorrect results).",
category=torch.jit.TracerWarning,
stacklevel=2,
)
return iter(self.unbind(0))
def __hash__(self):
# Do NOT handle __torch_function__ here as user's default
# implementation that handle most functions will most likely do it wrong.
# It can be easily overridden by defining this method on the user
# subclass if needed.
return id(self)
def __dir__(self):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__dir__, (self,), self)
tensor_methods = dir(self.__class__)
tensor_methods.remove("volatile") # deprecated
attrs = list(self.__dict__.keys())
keys = tensor_methods + attrs
# property only available dense, cuda tensors
if (not self.is_cuda) or self.is_sparse:
keys.remove("__cuda_array_interface__")
return sorted(keys)
# Numpy array interface, to support `numpy.asarray(tensor) -> ndarray`
__array_priority__ = 1000 # prefer Tensor ops over numpy ones
def __array__(self, dtype=None):
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__array__, (self,), self, dtype=dtype)
if dtype is None:
return self.numpy()
else:
return self.numpy().astype(dtype, copy=False)
# Wrap Numpy array again in a suitable tensor when done, to support e.g.
# `numpy.sin(tensor) -> tensor` or `numpy.greater(tensor, 0) -> ByteTensor`
def __array_wrap__(self, array):
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.__array_wrap__, (self,), self, array=array
)
if array.dtype == bool:
# Workaround, torch has no built-in bool tensor
array = array.astype("uint8")
return torch.from_numpy(array)
def __contains__(self, element: Any, /) -> bool:
r"""Check if `element` is present in tensor
Args:
element (Tensor or scalar): element to be checked
for presence in current tensor"
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__contains__, (self,), self, element)
if isinstance(
element, (torch.Tensor, Number, torch.SymInt, torch.SymFloat, torch.SymBool)
):
# type hint doesn't understand the __contains__ result array
return bool((element == self).any().item()) # type: ignore[union-attr]
raise RuntimeError(
f"Tensor.__contains__ only supports Tensor or scalar, but you passed in a {type(element)}."
)
@property
def __cuda_array_interface__(self):
"""Array view description for cuda tensors.
See:
https://numba.pydata.org/numba-doc/latest/cuda/cuda_array_interface.html
"""
if has_torch_function_unary(self):
# TODO mypy doesn't support @property, see: https://github.com/python/mypy/issues/6185
return handle_torch_function(
Tensor.__cuda_array_interface__.__get__, # type: ignore[attr-defined]
(self,),
self,
)
# raise AttributeError for unsupported tensors, so that
# hasattr(cpu_tensor, "__cuda_array_interface__") is False.
if not self.is_cuda:
raise AttributeError(
f"Can't get __cuda_array_interface__ on non-CUDA tensor type: {self.type()} "
"If CUDA data is required use tensor.cuda() to copy tensor to device memory."
)
if self.is_sparse:
raise AttributeError(
f"Can't get __cuda_array_interface__ on sparse type: {self.type()} "
"Use Tensor.to_dense() to convert to a dense tensor first."
)
# RuntimeError, matching tensor.__array__() behavior.
if self.requires_grad:
raise RuntimeError(
"Can't get __cuda_array_interface__ on Variable that requires grad. "
"If gradients aren't required, use var.detach() to get Variable that doesn't require grad."
)
typestr = _dtype_to_typestr(self.dtype)
itemsize = self.element_size()
shape = tuple(self.shape)
if self.is_contiguous():
# __cuda_array_interface__ v2 requires the strides to be omitted
# (either not set or set to None) for C-contiguous arrays.
strides = None
else:
strides = tuple(s * itemsize for s in self.stride())
data_ptr = self.data_ptr() if self.numel() > 0 else 0
data = (data_ptr, False) # read-only is false
return dict(typestr=typestr, shape=shape, strides=strides, data=data, version=2)
def storage_type(self):
r"""storage_type() -> type
Returns the type of the underlying storage.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.storage_type, (self,), self)
torch.storage._warn_typed_storage_removal()
return self._typed_storage()._get_legacy_storage_class()
def refine_names(self, *names):
r"""Refines the dimension names of :attr:`self` according to :attr:`names`.
Refining is a special case of renaming that "lifts" unnamed dimensions.
A ``None`` dim can be refined to have any name; a named dim can only be
refined to have the same name.
Because named tensors can coexist with unnamed tensors, refining names
gives a nice way to write named-tensor-aware code that works with both
named and unnamed tensors.
:attr:`names` may contain up to one Ellipsis (``...``).
The Ellipsis is expanded greedily; it is expanded in-place to fill
:attr:`names` to the same length as ``self.dim()`` using names from the
corresponding indices of ``self.names``.
Python 2 does not support Ellipsis but one may use a string literal
instead (``'...'``).
Args:
names (iterable of str): The desired names of the output tensor. May
contain up to one Ellipsis.
Examples::
>>> imgs = torch.randn(32, 3, 128, 128)
>>> named_imgs = imgs.refine_names('N', 'C', 'H', 'W')
>>> named_imgs.names
('N', 'C', 'H', 'W')
>>> tensor = torch.randn(2, 3, 5, 7, 11)
>>> tensor = tensor.refine_names('A', ..., 'B', 'C')
>>> tensor.names
('A', None, None, 'B', 'C')
.. warning::
The named tensor API is experimental and subject to change.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.refine_names, (self,), self, *names)
names = resolve_ellipsis(names, self.names, "refine_names")
return super().refine_names(names)
def align_to(self, *names):
r"""Permutes the dimensions of the :attr:`self` tensor to match the order
specified in :attr:`names`, adding size-one dims for any new names.
All of the dims of :attr:`self` must be named in order to use this method.
The resulting tensor is a view on the original tensor.
All dimension names of :attr:`self` must be present in :attr:`names`.
:attr:`names` may contain additional names that are not in ``self.names``;
the output tensor has a size-one dimension for each of those new names.
:attr:`names` may contain up to one Ellipsis (``...``).
The Ellipsis is expanded to be equal to all dimension names of :attr:`self`
that are not mentioned in :attr:`names`, in the order that they appear
in :attr:`self`.
Python 2 does not support Ellipsis but one may use a string literal
instead (``'...'``).
Args:
names (iterable of str): The desired dimension ordering of the
output tensor. May contain up to one Ellipsis that is expanded
to all unmentioned dim names of :attr:`self`.
Examples::
>>> tensor = torch.randn(2, 2, 2, 2, 2, 2)
>>> named_tensor = tensor.refine_names('A', 'B', 'C', 'D', 'E', 'F')
# Move the F and E dims to the front while keeping the rest in order
>>> named_tensor.align_to('F', 'E', ...)
.. warning::
The named tensor API is experimental and subject to change.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.align_to, (self,), self, *names)
ellipsis_idx = single_ellipsis_index(names, "align_to")
if ellipsis_idx is None:
return super().align_to(names)
return super().align_to(
[name for name in names if not is_ellipsis(name)], ellipsis_idx
)
def unflatten(self, dim, sizes): # type: ignore[override]
r"""
unflatten(dim, sizes) -> Tensor
See :func:`torch.unflatten`.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.unflatten, (self,), self, dim, sizes)
if not sizes:
raise RuntimeError("unflatten: sizes must be non-empty")
names = None
if isinstance(sizes, OrderedDict) or (
isinstance(sizes, (tuple, list)) and isinstance(sizes[0], (tuple, list))
):
names, sizes = unzip_namedshape(sizes)
return super().unflatten(dim, sizes, names)
else:
return super().unflatten(dim, sizes)
def rename_(self, *names, **rename_map):
"""In-place version of :meth:`~Tensor.rename`."""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.rename_, (self,), self, *names, **rename_map
)
# Note [rename_ / rename API]
# The Python API for these is different from the C++ API. In Python:
# 1) tensor.rename(*names) takes a vararglist of names
# 2) tensor.rename(**rename_map) takes a map of names to rename.
# C++ is static, making it difficult to implement similar behavior.
return update_names(self, names, rename_map, inplace=True)
def rename(self, *names, **rename_map):
"""Renames dimension names of :attr:`self`.
There are two main usages:
``self.rename(**rename_map)`` returns a view on tensor that has dims
renamed as specified in the mapping :attr:`rename_map`.
``self.rename(*names)`` returns a view on tensor, renaming all
dimensions positionally using :attr:`names`.
Use ``self.rename(None)`` to drop names on a tensor.
One cannot specify both positional args :attr:`names` and keyword args
:attr:`rename_map`.
Examples::
>>> imgs = torch.rand(2, 3, 5, 7, names=('N', 'C', 'H', 'W'))
>>> renamed_imgs = imgs.rename(N='batch', C='channels')
>>> renamed_imgs.names
('batch', 'channels', 'H', 'W')
>>> renamed_imgs = imgs.rename(None)
>>> renamed_imgs.names
(None, None, None, None)
>>> renamed_imgs = imgs.rename('batch', 'channel', 'height', 'width')
>>> renamed_imgs.names
('batch', 'channel', 'height', 'width')
.. warning::
The named tensor API is experimental and subject to change.
"""
if has_torch_function_unary(self):
return handle_torch_function(
Tensor.rename, (self,), self, *names, **rename_map
)
# See Note [rename_ / rename API]
return update_names(self, names, rename_map, inplace=False)
def to_sparse_coo(self):
"""Convert a tensor to :ref:`coordinate format <sparse-coo-docs>`.
Examples::
>>> dense = torch.randn(5, 5)
>>> sparse = dense.to_sparse_coo()
>>> sparse._nnz()
25
"""
return self.to_sparse()
def dim_order(
self, *, ambiguity_check: Union[bool, list[torch.memory_format]] = False
):
"""
dim_order(ambiguity_check=False) -> tuple
Returns the uniquely determined tuple of int describing the dim order or
physical layout of :attr:`self`.
The dim order represents how dimensions are laid out in memory of dense tensors,
starting from the outermost to the innermost dimension.
Note that the dim order may not always be uniquely determined.
If `ambiguity_check` is True, this function raises a RuntimeError when the dim order cannot be uniquely determined;
If `ambiguity_check` is a list of memory formats, this function raises a RuntimeError when tensor can not be interpreted
into exactly one of the given memory formats, or it cannot be uniquely determined.
If `ambiguity_check` is False, it will return one of legal dim order(s) without checking its uniqueness.
Otherwise, it will raise TypeError.
Args:
ambiguity_check (bool or List[torch.memory_format]): The check method for ambiguity of dim order.
Examples::
>>> torch.empty((2, 3, 5, 7)).dim_order()
(0, 1, 2, 3)
>>> torch.empty((2, 3, 5, 7)).transpose(1, 2).dim_order()
(0, 2, 1, 3)
>>> torch.empty((2, 3, 5, 7), memory_format=torch.channels_last).dim_order()
(0, 2, 3, 1)
>>> torch.empty((1, 2, 3, 4)).dim_order()
(0, 1, 2, 3)
>>> try:
... torch.empty((1, 2, 3, 4)).dim_order(ambiguity_check=True)
... except RuntimeError as e:
... print(e)
The tensor does not have unique dim order, or cannot map to exact one of the given memory formats.
>>> torch.empty((1, 2, 3, 4)).dim_order(
... ambiguity_check=[torch.contiguous_format, torch.channels_last]
... ) # It can be mapped to contiguous format
(0, 1, 2, 3)
>>> try:
... torch.empty((1, 2, 3, 4)).dim_order(ambiguity_check="ILLEGAL")
... except TypeError as e:
... print(e)
The ambiguity_check argument must be a bool or a list of memory formats.
.. warning::
The dim_order tensor API is experimental and subject to change.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.dim_order, (self,), self)
if self.is_sparse:
raise AttributeError(
f"Can't get dim order on sparse type: {self.type()} "
"Use Tensor.to_dense() to convert to a dense tensor first."
)
# Sanity check ambiguity_check data types
if not isinstance(ambiguity_check, bool):
if not isinstance(ambiguity_check, list):
raise TypeError(
"The ambiguity_check argument must be a bool or a list of memory formats."
)
for memory_format in ambiguity_check:
if not isinstance(memory_format, torch.memory_format):
raise TypeError(
"The ambiguity_check argument must be a bool or a list of memory formats."
)
def invalid_unique_memory_format(tensor, valid_memory_formats):
"""
Returns True if the tensor cannot be uniquely mapped to any of the given memory formats, False otherwise.
"""
n_legality = 0
for memory_format in valid_memory_formats:
if tensor.is_contiguous(memory_format=memory_format):
n_legality += 1
return n_legality != 1
def has_multiple_dim_order(tensor):
"""
Returns True if there're multiple legal dim orders for given tensor, False otherwise.
The tensor is considered to have multiple legal dim orders if either of the following conditions is met:
* Singleton Dimensions: There's at least one singleteon dimension in the tensor.
Since their size is 1, they don't affect the memory offset (stride * index
is zero because index is always zero). Therefore, they can be placed anywhere
in the dimension order without changing how data is accessed.
* Same strides: Strides reflect how the tensor is stored in memory.
If any two dimensions have the same stride, swapping these dimensions won't
change how data is accessed, leading to multiple correct dimension orders.
"""
sizes = tensor.size()
strides = tensor.stride()
# Check if there are any duplicate strides
has_duplicate_strides = any(
earlier == later for earlier, later in zip(strides, strides[1:])
)
# Check if there are any singleton dimensions
has_singleton_dims = any(size == 1 for size in sizes)
return has_duplicate_strides or has_singleton_dims
valid_memory_formats = (
ambiguity_check if isinstance(ambiguity_check, list) else []
)
check_multiple_dim_order = (
ambiguity_check if isinstance(ambiguity_check, bool) else True
)
if (
check_multiple_dim_order and has_multiple_dim_order(self)
) and invalid_unique_memory_format(self, valid_memory_formats):
raise RuntimeError(
"The tensor does not have unique dim order, or cannot map to exact one of the given memory formats."
)
import torch._prims_common as utils
return tuple(utils.compute_elementwise_output_logical_to_physical_perm(self))
def _update_names(self, names, inplace):
if has_torch_function_unary(self):
return handle_torch_function(
Tensor._update_names, (self,), self, names, inplace
)
# See Note [rename_ / rename API]
if inplace:
return super().rename_(names)
else:
return super().rename(names)
@classmethod
def __torch_function__(cls, func, types, args=(), kwargs=None):
"""
This __torch_function__ implementation wraps subclasses such that
methods called on subclasses return a subclass instance instead of
a ``torch.Tensor`` instance.
One corollary to this is that you need coverage for torch.Tensor
methods if implementing __torch_function__ for subclasses.
We recommend always calling ``super().__torch_function__`` as the base
case when doing the above.
While not mandatory, we recommend making `__torch_function__` a classmethod.
"""
if kwargs is None:
kwargs = {}
if not all(issubclass(cls, t) for t in types):
return NotImplemented
with _C.DisableTorchFunctionSubclass():
ret = func(*args, **kwargs)
if func in get_default_nowrap_functions():
return ret
else:
return _convert(ret, cls)
__torch_dispatch__ = _C._disabled_torch_dispatch_impl
def __dlpack__(self, stream=None):
"""
Creates a DLpack `capsule https://data-apis.org/array-api/latest/design_topics/data_interchange.html#data-interchange`_
of the current tensor to be exported to other libraries.
This function will be called from the `from_dlpack` method
of the library that will consume the capsule. `from_dlpack` passes the current
stream to this method as part of the specification.
Args:
stream (integer or None): An optional Python integer representing a
pointer to a CUDA stream. The current stream is synchronized with
this stream before the capsule is created, and since the capsule
shares its storage with the tensor this make it safe to access from
both streams. If None or -1 is passed then no synchronization is performed.
If 1 (on CUDA) or 0 (on ROCM) then the default stream is used for
synchronization.
"""
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__dlpack__, (self,), self, stream)
# DLPack capsules can't capture all of PyTorch's semantics,
# so we prohibit exporting tensors that would lose their properties like
# requires_grad and having the conjugate bit set.
if self.requires_grad:
raise RuntimeError(
"Can't export tensors that require gradient, use tensor.detach()"
)
if self.is_conj():
raise RuntimeError("Can't export tensors with the conjugate bit set")
if self.layout != torch.strided:
raise RuntimeError(
"Can't export tensors with layout other than torch.strided"
)
if stream is not None and type(stream) is not int:
# Stream pointers in CUDA/ROCm are uniquely numbered and can
# be retrieved from their integer value.
raise TypeError("stream must be ``int`` or ``none``")
elif stream is not None and stream != -1:
if self.device.type == "cuda":
# NB: This logic handles the special case values for default
# streams and must be kept in sync with from_dlpack in
# torch/utils/dlpack.py
if stream == 1 and torch.version.hip is None:
stream = torch.cuda.default_stream()
elif stream == 0 and torch.version.hip is not None:
stream = torch.cuda.default_stream()
else:
stream = torch.cuda.ExternalStream(stream)
# Only synchronize on different streams
sync_stream = torch.cuda.current_stream()
if stream != sync_stream:
event = torch.cuda.Event()
event.record(sync_stream)
stream.wait_event(event)
if self.device.type == "xla":
import torch_xla
import torch_xla.utils.dlpack as xla_dlpack
if (
len(torch_xla.real_devices()) <= 0
or "cuda" not in torch_xla.real_devices()[0].lower()
):
raise RuntimeError(
"Can't export to dlpack an XLA tensor that is not on CUDA."
)
return xla_dlpack.to_dlpack(self)
return torch.to_dlpack(self)
def __dlpack_device__(self) -> tuple[enum.IntEnum, int]:
if has_torch_function_unary(self):
return handle_torch_function(Tensor.__dlpack_device__, (self,), self)
from torch.utils.dlpack import DLDeviceType
device = self.device
idx = device.index if device.index is not None else 0
torch_device_type = device.type
if torch_device_type == "cuda" and torch.version.hip is not None:
device_type = DLDeviceType.kDLROCM
elif torch_device_type == "cpu" and self.is_pinned():
device_type = DLDeviceType.kDLCPUPinned
elif torch_device_type == "cuda":
device_type = DLDeviceType.kDLGPU
elif torch_device_type == "cpu":
device_type = DLDeviceType.kDLCPU
elif torch_device_type == "xpu":
device_type = DLDeviceType.kDLOneAPI
elif self.device.type == "privateuse1":
device_type = DLDeviceType.kDLExtDev
elif torch_device_type == "xla":
import torch_xla
if (
len(torch_xla.real_devices()) <= 0
or "cuda" not in torch_xla.real_devices()[0].lower()
):
raise ValueError(f"Unknown device type {torch_device_type} for Dlpack")
device_type = DLDeviceType.kDLGPU
else:
raise ValueError(f"Unknown device type {torch_device_type} for Dlpack")
return (device_type, idx)
__module__ = "torch"
def _convert(ret, cls):
if cls is Tensor:
return ret
if isinstance(ret, Tensor) and not isinstance(ret, cls):
ret = ret.as_subclass(cls)
if isinstance(ret, (tuple, list)):
# Also handles things like namedtuples
ret = type(ret)(_convert(r, cls) for r in ret)
return ret
|