pytorch/torch
2024-09-18 04:47:51 +00:00
..
_awaits Remove unused imported names in python files (#134438) 2024-08-27 20:44:04 +00:00
_C [MTIA] Support torch.cuda.get_device_capability equivalent API on MTIA (#135889) 2024-09-17 17:42:56 +00:00
_C_flatbuffer
_custom_op
_decomp Revert "Add decomposition for permute_copy (#130944)" 2024-09-17 13:42:55 +00:00
_dispatch
_dynamo [dynamo] Fix support for classmethod(property(...)) (#134968) 2024-09-18 04:47:51 +00:00
_export [export] Deserialize args with python keyword names (#136036) 2024-09-17 18:13:14 +00:00
_functorch [Traceable FSDP2][Partitioner] Must save AC output if output has a backward hook (#135727) 2024-09-14 08:45:58 +00:00
_higher_order_ops [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
_inductor [AO][Inductor] Enable WOQ fusion pattern with permute (#135928) 2024-09-18 00:56:16 +00:00
_lazy Remove unused imported names in python files (#134438) 2024-08-27 20:44:04 +00:00
_library [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
_logging Add torch._logging.scribe (#135224) 2024-09-05 22:37:13 +00:00
_numpy
_prims [hop] require hops to override __call__. (#134352) 2024-08-28 19:56:40 +00:00
_prims_common Allow cross-device copies for cpu scalars in refs (#135140) 2024-09-05 19:08:48 +00:00
_refs Revert "Add decomposition for permute_copy (#130944)" 2024-09-17 13:42:55 +00:00
_strobelight [Pytorch] Cleanup Strobelight URL and shorten for readability (#136102) 2024-09-16 18:10:33 +00:00
_subclasses [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
_vendor
amp [MPS] Add support for autocast in MPS (#99272) 2024-09-05 23:23:17 +00:00
ao [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
autograd Revert "fix compiled_autograd deadlock throw (#135795)" 2024-09-16 23:59:56 +00:00
backends [sparse] Add cuSPARSELt as a backend (#128534) 2024-08-21 22:06:07 +00:00
compiler [dynamo] simplify polyfill registration for builtins.all and builtins.any (#133769) 2024-08-29 20:56:16 +00:00
contrib
cpu Revise CPU vectorization ISA support API (#135075) 2024-09-05 12:14:56 +00:00
csrc [c10d] remove sleep from watchdogHandler (#135760) 2024-09-18 00:55:01 +00:00
cuda [amdsmi][torch] Update amdsmi API usages (#135504) 2024-09-10 19:15:39 +00:00
distributed Update real device in FSDP state_dict_utils (#134994) 2024-09-17 04:39:08 +00:00
distributions [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
export [TorchRec][PT2 IR][APF] short circuit the flatten/unflatten between EBC and KTRegroupAsDict modules (#136045) 2024-09-17 18:42:56 +00:00
fft
func
futures
fx use a fast expand algorithm (#135999) 2024-09-14 23:09:34 +00:00
jit [BE]: Update ruff to 0.6.0 (#133609) 2024-08-16 14:11:01 +00:00
legacy
lib
linalg Corrected docstring of `solve_triangular` (#129766) 2024-09-02 13:30:30 +00:00
masked [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
monitor
mps
mtia [MTIA] Support torch.cuda.get_device_capability equivalent API on MTIA (#135889) 2024-09-17 17:42:56 +00:00
multiprocessing [torch/multiprocessing] Use multiprocessing.reduction.register ForkingPickler.register to register custom tensor and storage reductions (#135030) 2024-09-16 20:07:29 +00:00
nested Support rms_norm() for NJT (#135872) 2024-09-17 18:09:20 +00:00
nn [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
onnx [ONNX] Fix numpy method to return the correct type (#136162) 2024-09-17 15:51:00 +00:00
optim Add back optim type hints that were lost when *.pyi files were removed (#136185) 2024-09-17 15:45:15 +00:00
package [3.13] fix 3.13 pickle error in torch/package (#136049) 2024-09-14 14:28:09 +00:00
profiler [Profiler] Torch Profiler distributed info is not JSON serializable (#135548) 2024-09-13 02:22:33 +00:00
quantization
signal Fix docstring for torch.signal.windows.nuttall (#134512) 2024-08-26 22:51:43 +00:00
sparse Add scaling arguments to bsr_dense_addmm (#136104) 2024-09-16 20:26:54 +00:00
special
testing Support rms_norm() for NJT (#135872) 2024-09-17 18:09:20 +00:00
utils [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
xpu [Intel GPU] Add XPU memory-related APIs (#129919) 2024-09-07 11:15:17 +00:00
__config__.py
__future__.py
__init__.py [Doc] update max-autotune for CPU (#134986) 2024-09-07 13:42:40 +00:00
_appdirs.py
_classes.py
_compile.py
_custom_ops.py
_deploy.py Remove unused imported names in python files (#134438) 2024-08-27 20:44:04 +00:00
_guards.py [PT2/Profiler] Add Context Info to Torch-Compiled Regions (#132765) 2024-08-27 04:55:04 +00:00
_jit_internal.py
_linalg_utils.py
_lobpcg.py [BE][Easy] enable ruff rule PIE790: unnecessary pass statement (#133200) 2024-08-15 15:50:19 +00:00
_lowrank.py
_meta_registrations.py [BE][Ez]: Add full half/bfloat16 dtype for unique and isin (#136114) 2024-09-16 17:49:12 +00:00
_namedtensor_internals.py
_ops.py [hop] require hops to override __call__. (#134352) 2024-08-28 19:56:40 +00:00
_python_dispatcher.py
_size_docs.py
_sources.py
_storage_docs.py
_streambase.py
_tensor_docs.py
_tensor_str.py
_tensor.py [BE]: Update mypy to 1.11.2 (#133816) 2024-09-16 19:44:11 +00:00
_torch_docs.py Fix/torch cat doc attr (#135698) 2024-09-11 22:32:55 +00:00
_utils_internal.py Revert "[Pytorch] Consolidate Strobelight compile time profiler between OSS and fbcode (#135953)" 2024-09-15 05:32:38 +00:00
_utils.py Add torch.serialization.skip_data context manager (#134504) 2024-09-05 16:53:39 +00:00
_VF.py Clean up RemoteCache classes (#134032) 2024-08-31 20:18:59 +00:00
_vmap_internals.py
_weights_only_unpickler.py Clarify error messages for NEWOBJ and BUILD in weights_only unpickler (#134346) 2024-08-27 14:45:39 +00:00
abi-check.cpp
CMakeLists.txt [split build] fix distributed problems (#134502) 2024-08-27 21:12:58 +00:00
custom_class_detail.h Enable clang-tidy coverage on torch/*.h (#133422) 2024-08-15 18:52:08 +00:00
custom_class.h Enable clang-tidy coverage on torch/*.h (#133422) 2024-08-15 18:52:08 +00:00
extension.h
functional.py
hub.py torch.hub: add get_dir/set_dir type hints (#134906) 2024-09-12 03:53:29 +00:00
library.h Enable clang-tidy coverage on torch/*.h (#133422) 2024-08-15 18:52:08 +00:00
library.py Improve opcheck docs. (#134692) 2024-08-28 22:22:04 +00:00
overrides.py Change wrapped_linear_prepack and wrapped_quantized_linear_prepacked to private by adding _ as prefix (#135401) 2024-09-08 04:16:24 +00:00
py.typed
quasirandom.py
random.py
README.txt
return_types.py
script.h
serialization.py [3.13] fix 3.13 pickle error in serialization.py (#136034) 2024-09-14 00:02:40 +00:00
storage.py Add torch.serialization.skip_data context manager (#134504) 2024-09-05 16:53:39 +00:00
torch_version.py
types.py [BE] better type annotation for torch.types (#129559) 2024-09-02 15:35:32 +00:00
version.py.tpl

Note [TH abstraction violation]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

TH/THC provide some hpp headers, which are proper C++ headers rather than
C headers.  These headers serve double duty as *internal implementation
detail* headers, whose contents should largely not be used by external
clients.

Ideally, we would not install these headers at all; instead, you should
use public functions (in headers like `THTensor.h`, NOT `THTensor.hpp`)
to manipulate these structs.  However, there are a few places
in torch/csrc where we violate this abstraction.  They are marked with
a pointer to this note.  Each of those sites will have to be refactored
when we refactor the guts of THTensor and related structures.