pytorch/torch
2023-03-03 23:40:00 +00:00
..
_awaits [jit] Support Awaitable type (#90863) 2023-01-30 17:38:59 +00:00
_C Fix typos under torch/_C directory (#95710) 2023-03-02 17:29:38 +00:00
_C_flatbuffer
_decomp fix nll loss decomposition to properly ignore ignore_index (#95833) 2023-03-02 08:37:56 +00:00
_dispatch
_dynamo Add export docs, improve asserts (#94961) 2023-03-03 23:40:00 +00:00
_export Add experimental torch.export prototype (#95070) 2023-02-28 02:40:19 +00:00
_functorch Add support for Inductor + symbolic shapes + training (#93059) 2023-02-28 22:44:31 +00:00
_inductor [inductor] use FP div in CPP expr printer (#95698) 2023-03-03 20:38:18 +00:00
_lazy
_prims Add support for nonzero, some improvements to reduce guards (#95387) 2023-02-24 00:27:45 +00:00
_prims_common Add support for nonzero, some improvements to reduce guards (#95387) 2023-02-24 00:27:45 +00:00
_refs [inductor] do not use ceil in arange ref (#95773) 2023-03-03 20:38:18 +00:00
_subclasses Don't use guardless contiguity/stride-like implementations (#95733) 2023-03-03 21:56:41 +00:00
amp [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
ao [Quant][PT2.0] fix issues for rearranging weight observer for decomposed linear (#94296) 2023-03-03 15:54:11 +00:00
autograd Implement sparse semantics support in gradcheck (2nd try) (#95405) 2023-02-27 17:48:02 +00:00
backends [MPS] Add optional minor argument to is_macos13_or_newer (#95065) 2023-02-17 18:30:20 +00:00
contrib
cpu
csrc [static-runtime] increase verbosity for schema check (#95937) 2023-03-03 06:50:28 +00:00
cuda memory viz: Add colors for categories and a legend (#90587) 2023-03-03 20:42:22 +00:00
distributed [DTensor][Shampoo] add _tenso.zero function (#95863) 2023-03-03 19:36:44 +00:00
distributions [BE] Remove dependency on six and future (#94709) 2023-02-14 09:14:14 +00:00
fft
func [functorch] linearize (#94173) 2023-02-09 15:45:08 +00:00
futures
fx Don't use guardless contiguity/stride-like implementations (#95733) 2023-03-03 21:56:41 +00:00
jit [2/3] Update .pyi Python stub files: Prettify rnn.py by using type annotated NamedTuple (#95267) 2023-03-01 19:37:23 +00:00
legacy
lib Simplify cmake code (#91546) 2023-02-08 01:05:19 +00:00
linalg
masked std/var: support floating point correction value (#94073) 2023-02-23 05:50:45 +00:00
monitor
mps [MPS] Enable Memory Leak Detection for test_mps.py (#94646) 2023-02-13 17:56:24 +00:00
multiprocessing [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
nested
nn [3/3] Update .pyi Python stub files and enable 'UFMT' linter (#95268) 2023-03-01 23:50:56 +00:00
onnx Apply peephole for eval mode when constant folding is enabled only (#95801) 2023-03-01 23:07:38 +00:00
optim [optim] Widen the cases for defaulting to foreach (#95820) 2023-03-02 04:15:33 +00:00
package [BE] Add flake8-logging-format linter (#94840) 2023-02-15 17:54:50 +00:00
profiler [profiler] update docs with repeat=1 (#95085) 2023-02-21 21:09:10 +00:00
quantization AO migration: replace torch internal callsites (#94170) 2023-02-07 02:32:23 +00:00
signal Reland "Add torch.utils.device_mode" (#91796) 2023-01-09 20:57:12 +00:00
sparse Triton kernel for bsr @ dense (#94823) 2023-03-03 15:11:28 +00:00
special
testing [MTPG] Improve all_reduce and handle bwd thread support (#95524) 2023-03-03 18:53:36 +00:00
utils Fix: validate_input_col for partial functions (#95067) 2023-03-03 21:05:07 +00:00
__config__.py
__future__.py
__init__.py Automatically guard when SymInt is converted to int (#95479) 2023-02-25 19:41:51 +00:00
_appdirs.py
_classes.py [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
_deploy.py
_guards.py [BE] Simplify Source.is_nn_module; add some types (#95292) 2023-02-22 22:33:58 +00:00
_jit_internal.py [JIT] Improve source attribution for NamedTuple type inference (#95761) 2023-03-01 23:40:13 +00:00
_linalg_utils.py Remove deprecated torch.symeig (#70988) 2023-01-31 11:59:11 +00:00
_lobpcg.py [BE]: pyupgrade Python to 3.8 - imports and object inheritance only (#94308) 2023-02-07 21:10:56 +00:00
_lowrank.py
_meta_registrations.py Fixes for PyTorch/XLA functionalization integration (#94537) 2023-03-02 23:02:34 +00:00
_namedtensor_internals.py
_ops.py [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
_python_dispatcher.py
_sources.py [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
_storage_docs.py
_tensor_docs.py Add further info to masked_scatter and masked_scatter_ documention (#94545) 2023-02-15 07:50:47 +00:00
_tensor_str.py [BE] Remove dependency on six and future (#94709) 2023-02-14 09:14:14 +00:00
_tensor.py Add HPU to compatible shallow copy list and remove lazy HPU changes (#94673) 2023-02-14 17:15:25 +00:00
_torch_docs.py docs: Match open bracket with close bracket in unsqueeze (#95215) 2023-02-24 03:56:59 +00:00
_utils_internal.py
_utils.py [BE]: pyupgrade Python to 3.8 - imports and object inheritance only (#94308) 2023-02-07 21:10:56 +00:00
_VF.py [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
_vmap_internals.py
_weights_only_unpickler.py Add float to list of allowed ops (#94910) 2023-02-15 23:13:21 +00:00
abi-check.cpp
CMakeLists.txt fix some tiny code issues (#95757) 2023-03-01 23:27:32 +00:00
custom_class_detail.h More tidy fixes (#93069) 2023-01-27 06:40:50 +00:00
custom_class.h More fixes and improved clang-tidy checkers (#93213) 2023-02-01 14:44:17 +00:00
extension.h
functional.py Add dimension check in tensordot (#94497) 2023-03-02 05:45:11 +00:00
hub.py [BE]: pyupgrade Python to 3.8 - imports and object inheritance only (#94308) 2023-02-07 21:10:56 +00:00
library.h
library.py
overrides.py Reland "Add torch.empty_permuted (#95069)" (#95208) 2023-02-21 18:02:48 +00:00
py.typed
quasirandom.py [BE]: pyupgrade Python to 3.8 - imports and object inheritance only (#94308) 2023-02-07 21:10:56 +00:00
random.py [MPS] Add Python Module Bindings for the MPS backend (#94417) 2023-02-12 21:22:30 +00:00
README.txt
return_types.py
script.h
serialization.py [BE] Remove dependency on six and future (#94709) 2023-02-14 09:14:14 +00:00
storage.py [BE] [2/3] Rewrite super() calls in functorch and torch (#94588) 2023-02-10 21:16:33 +00:00
torch_version.py
types.py [BE]: pyupgrade Python to 3.8 - imports and object inheritance only (#94308) 2023-02-07 21:10:56 +00:00

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.