frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-11-12 06:44:55 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
bdb225e9f0c15f968235c09b2f14b3b5c6f343f7
pytorch/torch
History
Mike Ruberry bdb225e9f0 Revert D27478436: Use tensorpipe::Buffer::device() instead of tensorpipe::Buffer::deviceType(). 
Test Plan: revert-hammer

Differential Revision:
D27478436 (3e185253b6)

Original commit changeset: 3962257bc623

fbshipit-source-id: 6619617af2b32445473f21e73bf3841dd7a491b2
2021-04-02 09:08:50 -07:00
..
_C
Lint trailing newlines (#54737)
2021-03-30 13:09:52 -07:00
autograd
Refactor get numerical jacobian (#54092)
2021-03-31 16:28:16 -07:00
backends
Add size op in nnapi serializer (#52026)
2021-02-10 15:57:01 -08:00
contrib
Add type annotations to _tensorboard_vis.py and hipify_python.py (#49834)
2021-01-04 09:29:51 -08:00
csrc
Revert D27478436: Use tensorpipe::Buffer::device() instead of tensorpipe::Buffer::deviceType().
2021-04-02 09:08:50 -07:00
cuda
Back out "[pytorch][PR] Merge CUDA Streams and Events" (#54996)
2021-03-31 10:21:35 -07:00
distributed
[Gradient Compression] Report compression rate for batched PowerSGD hook (#55103)
2021-03-31 22:17:05 -07:00
distributions
Remove legacy constructor calls from _torch_ folder. (#53889)
2021-03-19 15:20:19 -07:00
fft
[doc] Fix documentations of torch functions (#52982)
2021-03-01 09:59:57 -08:00
for_onnx
Ignore F401 in all __init__.py without putting noqa (#25823)
2019-10-23 15:28:13 -07:00
futures
Ensure torch.futures.wait_all exits early on error. (#53953)
2021-03-25 07:39:14 -07:00
fx
[FX] Adds C-level monkeypatching of torch.randn so that we can capture it during tracing. (#54060)
2021-04-01 07:34:31 -07:00
jit
Back out "[pytorch][PR] Merge CUDA Streams and Events" (#54996)
2021-03-31 10:21:35 -07:00
legacy
Remove torch/legacy (#11823)
2018-09-20 14:00:54 -07:00
lib
[DDP Logging] Log use of uneven inputs API (#54919)
2021-04-01 16:22:32 -07:00
linalg
SVD docs improved (#54002)
2021-04-01 09:31:40 -07:00
multiprocessing
Make SharedCache thread-safe (#53750)
2021-03-25 06:35:03 -07:00
nn
[DDP Logging] Log use of uneven inputs API (#54919)
2021-04-01 16:22:32 -07:00
onnx
[ONNX] Fix export of copy_ operator (#51938) (#54870)
2021-03-31 21:14:32 -07:00
optim
Lint trailing newlines (#54737)
2021-03-30 13:09:52 -07:00
package
[package] rename PackageExporter.external to PacakgeExporter.extern_modules (#54601)
2021-03-25 11:50:07 -07:00
profiler
Add FLOPS support to the new profiler API. (#51734)
2021-02-05 15:03:35 -08:00
quantization
Revert D27408378: [quant][graphmode][fx][refactor] Factor out insert_observers_for_model to a separate function
2021-03-31 20:51:42 -07:00
sparse
Forbid trailing whitespace (#53406)
2021-03-05 17:22:55 -08:00
special
[special] Alias for special.expm1 and special.exp2 (#54670)
2021-03-30 10:03:13 -07:00
testing
Port torch.copysign method_tests() to OpInfo (#54945)
2021-04-01 20:30:25 -07:00
utils
docs: fix docstring signature of torch.{onnx,utils} (#54662)
2021-03-29 10:07:42 -07:00
__config__.py
Expose CXX_FLAGS through __config__ (#47861)
2020-12-01 19:58:29 -08:00
__future__.py
Add torch.__future__._overwrite_module_params_on_conversion global flag, and check it in nn.Module._apply() (#21613)
2019-06-19 10:30:02 -07:00
__init__.py
[codemod] fix tautological imports
2021-03-27 01:15:57 -07:00
_appdirs.py
Delete Python <= 3.5 specific checks from the code (#39879)
2020-06-15 08:16:06 -07:00
_autograd_functions.py
lu_backward: more numerically stable and with complex support. (#53994)
2021-03-25 13:33:58 -07:00
_classes.py
[*.py] Rename "Arguments:" to "Args:" (#49736)
2020-12-28 09:34:47 -08:00
_deploy.py
[package] Pull out _UnpicklerWrapper into PackageUnpickler (#53049)
2021-03-01 18:40:52 -08:00
_jit_internal.py
[package] fix mangling issues with TorchScript (#54915)
2021-03-31 00:58:05 -07:00
_linalg_utils.py
Drop unused imports (#49972)
2021-01-13 12:26:17 -08:00
_lobpcg.py
[*.py] Rename "Arguments:" to "Args:" (#49736)
2020-12-28 09:34:47 -08:00
_lowrank.py
Drop unused imports (#49972)
2021-01-13 12:26:17 -08:00
_namedtensor_internals.py
Remove (most) Python 2 support from Python code (#35615)
2020-04-22 09:23:14 -07:00
_ops.py
Back out "Revert D26077905: Back out "Revert D25850783: Add torch::deploy, an embedded torch-python interpreter"" (#51267)
2021-01-28 19:30:45 -08:00
_python_dispatcher.py
Rename DefaultBackend to CompositeExplicitAutograd (#54470)
2021-03-26 10:53:30 -07:00
_six.py
Clean up usage of torch._six partially (#49785)
2021-02-08 13:58:34 -08:00
_storage_docs.py
Python API for Complex Storage and storage copy logic (#35771)
2020-05-01 11:47:22 -07:00
_tensor_docs.py
docs: remove extra backticks in narrow_copy (#54669)
2021-03-29 10:38:21 -07:00
_tensor_str.py
Reland: Add base forward grad logic (#49734)
2020-12-22 12:11:27 -08:00
_tensor.py
Fix override for __iter__ (#54702)
2021-03-30 08:30:50 -07:00
_torch_docs.py
SVD docs improved (#54002)
2021-04-01 09:31:40 -07:00
_utils_internal.py
Back out "Revert D26077905: Back out "Revert D25850783: Add torch::deploy, an embedded torch-python interpreter"" (#51267)
2021-01-28 19:30:45 -08:00
_utils.py
Back out "[pytorch][PR] Merge CUDA Streams and Events" (#54996)
2021-03-31 10:21:35 -07:00
_VF.py
Address JIT/Mypy issue with torch._VF (#43454)
2020-08-25 09:23:54 -07:00
_vmap_internals.py
Beef up {jacobian, hessian} vectorize docs; eliminate a warning (#51638)
2021-02-03 17:15:16 -08:00
abi-check.cpp
Fixes for Torch Script C++ API (#11682)
2018-09-17 09:54:50 -07:00
CMakeLists.txt
Link onnx_library when BUILD_TEST=0 for Windows (#51937)
2021-03-31 10:58:25 -07:00
custom_class_detail.h
[PyTorch] Remove reference_cast in make_boxed_from_unboxed_functor (#51319)
2021-02-17 10:58:44 -08:00
custom_class.h
[PyTorch][easy] Move more strings in torch::class_ (#54547)
2021-03-29 12:27:11 -07:00
deploy.h
Lint trailing newlines (#54737)
2021-03-30 13:09:52 -07:00
extension.h
Remove deprecated variable_tensor_functions (#15003)
2018-12-11 17:16:11 -08:00
functional.py
Deprecate torch.chain_matmul in favor of torch.linalg.multi_dot (#53453)
2021-04-01 04:50:51 -07:00
hub.py
Only allow hub.load() from original repo. (#54451)
2021-03-22 20:27:54 -07:00
library.h
Make meta a device (getting rid of empty_meta) (#53143)
2021-03-03 11:24:13 -08:00
overrides.py
Deprecate torch.chain_matmul in favor of torch.linalg.multi_dot (#53453)
2021-04-01 04:50:51 -07:00
py.typed
remediation of S205607
2020-07-17 17:19:47 -07:00
quasirandom.py
[SobolEngine] Fix edge case of dtype of first sample (#51578)
2021-02-02 14:24:56 -08:00
random.py
[*.py] Rename "Arguments:" to "Args:" (#49736)
2020-12-28 09:34:47 -08:00
README.txt
…
script.h
Implement public API InferenceMode and its error handling (#55008)
2021-03-31 10:48:00 -07:00
serialization.py
docs: reference links to serialization.html (#54659)
2021-03-29 10:15:07 -07:00
storage.py
Add type informations to torch/storage.py (#46876)
2020-11-06 11:34:10 -08:00
types.py
Enable torch.tensor typechecks (#45077)
2020-09-24 08:22:06 -07:00

README.txt 

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.