pytorch/torch
Chen Lai cee84f4051 fix model dump for the lowered module (#72866)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/72866

https://github.com/pytorch/pytorch/pull/71597 adds a wrapper `torch.jit.LoweredWrapper` and it breaks the model dump. Fix the model_dump in the notebook
ghstack-source-id: 149311636

Test Plan:
CI and test with N509022

Before:

{F701413403}

After:

{F701412963}

Reviewed By: iseeyuan

Differential Revision: D34247216

fbshipit-source-id: 695b02b03675fae596bb450441b327e4cdcffe9c
(cherry picked from commit d46a82a4c125722ca17f67b576b2aec523de1f15)
2022-02-17 07:09:44 +00:00
..
_C Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
_masked Generate static docstrings for torch._masked functions. (#72865) 2022-02-17 02:44:16 +00:00
ao Revert D34126542: [Qunat] Add ConvTranspose reference module 2022-02-16 22:24:15 +00:00
autograd Add autograd tests for addmm, addmv, mm, mv and CSR matrix input (#71949) 2022-02-11 23:14:02 +00:00
backends Fix type annotation for torch.backends.cudnn.allow_tf32 (#72757) 2022-02-14 17:26:37 +00:00
contrib
cpu
csrc [Static Runtime] Remove ProcessedNode::num_outputs_ (#72592) 2022-02-17 05:09:17 +00:00
cuda Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
distributed [PT-D][Sharded Tensor] new init api for local tensor and sharding spec auto inference (#72733) 2022-02-16 17:42:39 +00:00
distributions Revert D34245091: [pytorch][PR] Improve numerical stability of torch.distributions.wishart.Wishart 2022-02-16 21:40:07 +00:00
fft
for_onnx
futures
fx [graph_manipulation] Unpack list of outputs (#72940) 2022-02-17 06:38:52 +00:00
jit Make fusion strategy api public (#72639) 2022-02-16 03:45:15 +00:00
legacy
lib
linalg
monitor
multiprocessing Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
nn Add docstrings to native_channel_shuffle (#72919) 2022-02-17 02:33:08 +00:00
onnx fix: onnx PReLU unidirectional broadcasting 2022-02-16 22:28:08 +00:00
optim Add missing import (#72840) 2022-02-15 19:43:54 +00:00
package Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
profiler
quantization
sparse
special
testing only compare attributes for meta tensors (#72508) 2022-02-17 02:33:08 +00:00
utils fix model dump for the lowered module (#72866) 2022-02-17 07:09:44 +00:00
__config__.py
__future__.py
__init__.py Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
_appdirs.py
_classes.py
_deploy.py Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
_jit_internal.py (2/2) Make TorchScript Preserve Fully Qualified Class Name for Python Exceptions: frontend change (#72899) 2022-02-16 03:45:15 +00:00
_linalg_utils.py
_lobpcg.py
_lowrank.py
_namedtensor_internals.py
_ops.py Add torch.ops per overload API (#72206) 2022-02-11 17:19:06 +00:00
_python_dispatcher.py Revert D34034848: free up dispatch key space (in C++) 2022-02-14 23:29:00 +00:00
_six.py
_sources.py
_storage_docs.py
_tensor.py Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
_tensor_docs.py Revert D33800694: [pytorch][PR] scatter_reduce documentation 2022-02-15 20:10:26 +00:00
_tensor_str.py
_torch_docs.py Revert D33800694: [pytorch][PR] scatter_reduce documentation 2022-02-15 20:10:26 +00:00
_utils.py Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
_utils_internal.py
_VF.py
_vmap_internals.py
abi-check.cpp
autocast_mode.py
CMakeLists.txt Allow caffe2-specific graph transformations for OperatorExportTypes.ONNX_ATEN_FALLBACK when BUILD_CAFFE2 is ON (#67460) (#68490) 2022-02-10 03:26:48 +00:00
custom_class.h
custom_class_detail.h
deploy.h
extension.h
functional.py
hub.py
library.h
overrides.py avoiding adding some functions to the public python API before 1.11 release (#72543) 2022-02-14 19:49:01 +00:00
py.typed
quasirandom.py
random.py
README.txt
return_types.py
script.h
serialization.py Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
storage.py Rename Typed/UntypedStorage to _Typed/_UntypedStorage (#72540) 2022-02-15 23:53:01 +00:00
torch_version.py
types.py

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.