Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/73284
Some important ops won't support optional type until opset 16,
so we can't fully test things end-to-end, but I believe this should
be all that's needed. Once ONNX Runtime supports opset 16,
we can do more testing and fix any remaining bugs.
Test Plan: Imported from OSS
Reviewed By: albanD
Differential Revision: D34625646
Pulled By: malfet
fbshipit-source-id: 537fcbc1e9d87686cc61f5bd66a997e99cec287b
Co-authored-by: BowenBao <bowbao@microsoft.com>
Co-authored-by: neginraoof <neginmr@utexas.edu>
Co-authored-by: Nikita Shulga <nshulga@fb.com>
(cherry picked from commit 822e79f31ae54d73407f34f166b654f4ba115ea5)
Updating the docstrings and type annotations as I walk through the code.
- Turned some comments into docstrings.
- Added type annotations for some functions in utils and the registry
- Removed direct function imports; importing functions makes name space collision easier to happen and refactoring/code analysis harder: https://google.github.io/styleguide/pyguide.html#22-imports
- Formatted touched files with black
Pull Request resolved: https://github.com/pytorch/pytorch/pull/76255
Approved by: https://github.com/BowenBao
Currently `torch.onnx.export(.., operator_export_type=OperatorExportTypes.ONNX_ATEN_FALLBACK)` only issues ATen ops through explicit requests (e.g. `g.at()`) calls inside each op symbolic function. This is done based on specific conditions such as `operator_export_type==OperatorExportTypes.ONNX_ATEN_FALLBACK)` or `is_caffe2_aten_fallback()`
This PR extends the ATen fallback mechanism for scenarios when the symbolic function raises `RuntimeError` during export. The idea is that partial implementation of existing ONNX ops can fallback to ATen as a last resort. That is valuable because each operator can have many input combinations and not all are always implemented.
A minor fix was done to make sure the `overload_name` attribute is added to explicit ATen op fallback requests when a symbolic is not registered to a particular op.
ps: The behavior for builds with BUILD_CAFFE2=1 is not changed to ensure BC.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/74759
Approved by: https://github.com/garymm, https://github.com/msaroufim
Previously pre-tracing model is required for exporting quantized model.
e.g. calling `traced_m = torch.jit.trace(model, inputs)` and export `traced_m`.
The reason was quantized weights are stored in a unique `PackedParam` structure,
and they need to be handled by tracing to be exportable.
This PR enables export api to call tracing underneath if it detects quantization
in the model.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/75921
Approved by: https://github.com/garymm
This PR introduces 3 BC changes:
First, this PR propagates `BUILD_CAFFE2` flag to `libtorch` and `libtorch_python`, which is necessary for non-caffe2 ONNX runtimes when using `ONNX_ATEN_FALLBACK` operator export type.
Second, as a complement of https://github.com/pytorch/pytorch/pull/68490, this PR refactors Caffe2's Aten ops symbolics to consider not only the `operator_export_type` (aka `ONNX_ATEN_FALLBACK`) to emit Caffe2 Aten ops, but also whether `BUILD_CAFFE2` (which is called `torch.onnx._CAFFE2_ATEN_FALLBACK` in python binding) is set.
Lastly, it renames `onnx::ATen` to `aten::ATen` for ONNX spec consistency in a BC fashion.
ONNX doesn't have `ATen` op on its spec, but PyTorch ONNX converter emits them. Non-Caffe2 backend engines would be mislead by such operator's name/domain. A non-ideal workaround would be to have Aten ops handled based on its name and ignore the (non-complaint) domain. Moreover, users could incorrectly file bugs to either ONNX or ONNX Runtime when they inspect the model and notice the presence of an unspecified ONNX operator.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/73954
Approved by: https://github.com/BowenBao, https://github.com/malfet, https://github.com/garymm, https://github.com/jiafatom
Previous logic didn't consider the case for TrainingMode.PRESERVE.
A more direct way is to check `model.training`, which is the accurate
training mode, set by `exporter_context(model, training)`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/74247
Approved by: https://github.com/garymm
Summary:
Add ONNX exporter logging facility. Supporting both C++/Python logging api. Logging can be turned on/off. Logging output stream can be either set to `stdout` or `stderr`.
A few other changes:
* When exception is raised in passes, the current IR graph being processed will be logged.
* When exception is raised from `_jit_pass_onnx` (the pass that converts nodes from namespace `ATen` to `ONNX`), both ATen IR graph and ONNX IR graph under construction will be logged.
* Exception message for ConstantFolding is truncated to avoid being too verbose.
* Update the final printed IR graph with node name in ONNX ModelProto as node attribute. Torch IR Node does not have name. Adding this to printed IR graph helps debugging.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/71342
Reviewed By: msaroufim
Differential Revision: D34433473
Pulled By: malfet
fbshipit-source-id: 4b137dfd6a33eb681a5f2612f19aadf5dfe3d84a
(cherry picked from commit 67a8ebed5192c266f604bdcca931df6fe589699f)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/73280
This PR adds a new attribute overload_name to the Aten node so that third party applications can implement calls to libtorch without using PyTorch source code.
This is necessary because torch's torch::jit::findOperatorFor(fullname) requires a full name, including operator and overload names.
ATen op was originally created for Caffe2, which leveraged the availability of the pytorch yaml files to create calls to the aten oeprators directly, not relying on torch::jit::findOperatorFor
The first part of the PR refactors all symbolics that create Aten ops, so that there is a single helper for this operator. Next all symbolics are updated to pass in the relevant overload name, if empty string is not applicable
Test Plan: Imported from OSS
Reviewed By: jbschlosser
Differential Revision: D34625645
Pulled By: malfet
fbshipit-source-id: 37d58cfb5231833768172c122efc42edf7d8609a
(cherry picked from commit e92f09117d3645b38bc3235b30aba4b4c7c71dfa)
Enables local function export to capture annotated attributes.
For example:
```python
class M(torch.nn.Module):
num_layers: int
def __init__(self, num_layers):
super().__init__()
self.num_layers = num_layers
def forward(self, args):
...
```
`num_layers` will now be captured as attribute of local function `M`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/72883
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/69547
ScriptModule export introduces duplicated ONNX initializers for shared weights, unnecessarily increases ONNX model size. This PR de-duplicates ONNX initializers for model exported in eval mode, by checking if the underlying tensors share the same `data_ptr`, `strides` and `sizes`.
Test Plan: Imported from OSS
Reviewed By: msaroufim
Differential Revision: D32994271
Pulled By: malfet
fbshipit-source-id: 10ac66638b6255890875272472aa9ed07a5b1d9a
Co-authored-by: BowenBao <bowbao@microsoft.com>
(cherry picked from commit d7cbde940c)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/68491
* Allows implementing symbolic functions for domains other than `aten`, for example `prim`, in symbolic_opset#.py.
* Allows symbolic function to access extra context if needed, through `SymbolicFunctionState`.
* Particularly, the `prim::PythonOp` special case can access node without the need of passing node through inputs. Updates will be made downstreams, and in a follow-up PR we will remove the previous workaround in exporter.
* `prim::Loop`, `prim::If`, etc are now moved outside of `_run_symbolic_function` from utils.py, and to symbolic_opset9.py.
Motivation for this change:
- Better maintainability and reducing complexity. Easier to add symbolic for operators, both simple and complex ones (that need additional context), without the former needing to know the existence of the latter.
- The design idea was long outdated. prim ops are no longer rare special cases, and they shouldn't all be handled inside `_run_symbolic_function`. As a result this function becomes too clumsy. There were also prim ops symbolic added in symbolic_opset#.py with signature `prim_[opname]`, creating separation and confusion.
Test Plan: Imported from OSS
Reviewed By: jansel
Differential Revision: D32483782
Pulled By: malfet
fbshipit-source-id: f9affc31b1570af30ffa6668da9375da111fd54a
Co-authored-by: BowenBao <bowbao@microsoft.com>
(cherry picked from commit 1e04ffd2fd)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/68490
The use of ATEN as a fallback operator during ONNX conversion is important for increasing operator coverage or even provide more efficient implementations over some ONNX ops.
Currently this feature is available through `OperatorExportTypes.ONNX_ATEN_FALLBACK`,
but it also performs changes to the graph that are runnable by Caffe2, only.
This PR introduces restricts caffe2-specific graph transformations for `ONNX_ATEN_FALLBACK`
operator export type for when pytorch is built with caffe2 support (aka BUILD_CAFFE2=1 during build)
The first version of this PR introduced a new operator export type `ONNX_ATEN__STRICT_FALLBACK`,
which essentially is the same as `ONNX_ATEN_FALLBACK` but without caffe2 transformations.
It was preferred to not introduce a new operator export type, but to refine the existing aten fallback one
## BC-breaking note
### The global constant `torch.onnx.PYTORCH_ONNX_CAFFE2_BUNDLE` is removed in favor of
a less visible `torch.onnx._CAFFE2_ATEN_FALLBACK`.
`PYTORCH_ONNX_CAFFE2_BUNDLE` is really a dead code flag always set to False.
One alternative would be fixing it, but #66658 disables Caffe2 build by default.
Making a Caffe2 feature a private one seems to make more sense for future deprecation.
### The method `torch.onnx.export` now defaults to ONNX when `operator_export_type` is not specified.
Previously `torch.onnx.export's operator_export_type` intended to default to `ONNX_ATEN_FALLBACK` when `PYTORCH_ONNX_CAFFE2_BUNDLE` was set, but it would never happen as `PYTORCH_ONNX_CAFFE2_BUNDLE` is always undefined
Co-authored-by: Nikita Shulga <nshulga@fb.com>
Test Plan: Imported from OSS
Reviewed By: jansel
Differential Revision: D32483781
Pulled By: malfet
fbshipit-source-id: e9b447db9466b369e77d747188685495aec3f124
(cherry picked from commit 5fb1eb1b19)
Cover more cases of scope inferencing where consecutive nodes don't have valid scope information. Usually these nodes are created in some pass where authors forgot to assign meaningful scope to them.
* One rule of `InferScope` is to check if the current node's outputs' users share the same scope. Recursively run `InferScope` on the user nodes if they are missing scope as well. Since the graph is SSA, the depth is finite.
* Fix one pass that missed scope information for a new node.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/71897
Export should fail if export_modules_as_functions is set and opset_version<15.
This is because opeset_version < 15 implies IR version < 8, which means no local function support.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/71619
Summary:
For some ONNX exported models, the inputs/outputs names have sometimes a numeric value and this makes pretty hard to inspect the generated graphs in the case of large models.
The solution in this PR was initially submitted to our internal utilities library by take-cheeze https://github.com/pfnet/pytorch-pfn-extras/pull/102
Now we would like to upstream this change by adding an extra kwarg when exporting the model to allow replacing these numeric names with actual debuggable ones.
As an example, the following code shows that the module output is `3`
```python
g, p, o = _model_to_graph(module, torch.ones(1, 10))
for n in g.nodes():
for v in n.outputs():
print(v.debugName())
```
output
```
3
```
With this PR
```
v3_Gemm
```
This allows identifying this out as a value from the associated Gemm layer.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/68976
Reviewed By: jansel
Differential Revision: D33662246
Pulled By: msaroufim
fbshipit-source-id: 45f56eef2a84d9a318db20c6a6de6c2743b9cd99
(cherry picked from commit 513c1d28f1)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/69546
The arg is not used and was previously deprecated.
Also remove torch.onnx._export_to_pretty_string. It's redundant with the
public version.
Test Plan: Imported from OSS
Reviewed By: malfet
Differential Revision: D32994270
Pulled By: msaroufim
fbshipit-source-id: f8f3933b371a0d868d9247510bcd73c31a9d6fcc
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/67803
* Addresses comments from #63589
[ONNX] remove torch::onnx::PRODUCER_VERSION (#67107)
Use constants from version.h instead.
This simplifies things since we no longer have to update
PRODUCER_VERSION for each release.
Also add TORCH_VERSION to version.h so that a string is available for
this purpose.
[ONNX] Set `ir_version` based on opset_version. (#67128)
This increases the odds that the exported ONNX model will be usable.
Before this change, we were setting the IR version to a value which may
be higher than what the model consumer supports.
Also some minor clean-up in the test code:
* Fix string replacement.
* Use a temporary file so as to not leave files around in the test
current working directory.
Test Plan: Imported from OSS
Reviewed By: msaroufim
Differential Revision: D32181306
Pulled By: malfet
fbshipit-source-id: 02f136d34ef8f664ade0bc1985a584f0e8c2b663
Co-authored-by: BowenBao <bowbao@microsoft.com>
Co-authored-by: Gary Miguel <garymiguel@microsoft.com>
Co-authored-by: Nikita Shulga <nshulga@fb.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/66143
Delete test_list_remove. There's no point in testing conversion of
this model since TorchScript doesn't support it.
Add a link to an issue tracking test_embedding_bag_dynamic_input.
[ONNX] fix docs (#65379)
Mainly fix the sphinx build by inserting empty before
bulleted lists.
Also some minor improvements:
Remove superfluous descriptions of deprecated and ignored args.
The user doesn't need to know anything other than that they are
deprecated and ignored.
Fix custom_opsets description.
Make indentation of Raises section consistent with Args section.
[ONNX] publicize func for discovering unconvertible ops (#65285)
* [ONNX] Provide public function to discover all unconvertible ATen ops
This can be more productive than finding and fixing a single issue at a
time.
* [ONNX] Reorganize test_utility_funs
Move common functionality into a base class that doesn't define any
tests.
Add a new test for opset-independent tests. This lets us avoid running
the tests repeatedly for each opset.
Use simple inheritance rather than the `type()` built-in. It's more
readable.
* [ONNX] Use TestCase assertions rather than `assert`
This provides better error messages.
* [ONNX] Use double quotes consistently.
[ONNX] Fix code block formatting in doc (#65421)
Test Plan: Imported from OSS
Reviewed By: jansel
Differential Revision: D31424093
fbshipit-source-id: 4ced841cc546db8548dede60b54b07df9bb4e36e
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/66140
* Add new argument to export api to enable users specifying `nn.Module` classes that they wish to be exported as local function in ONNX model.
* Refactor `torch/csrc/jit/serialization/export.cpp`, and remove redundant `EncoderBase` class.
* ~~Contains changes from #63268~~
* Depends on #63716 to update onnx submodule.
Test Plan: Imported from OSS
Reviewed By: jansel
Differential Revision: D31424098
fbshipit-source-id: c949d0b01c206c30b4182c2dd1a5b90e32b7a0d3
Co-authored-by: BowenBao <bowbao@microsoft.com>
Summary:
Apex O2 hook state_dict to return fp16 weights as fp32. Exporter cannot identify them as same tensors.
Since this hook is only used by optimizer, it is safe to remove this hook while exporting.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/66700
Reviewed By: zou3519
Differential Revision: D31695132
Pulled By: malfet
fbshipit-source-id: 977bdf57240002498f3ad0f1a8046c352e9860e6
Summary:
This moves it to where the user would expect it to be based on the
documentation and all the other public classes in the torch.onnx module.
Also rename it from ONNXCheckerError, since the qualified name
torch.onnx.ONNXCheckerError is otherwise redundant.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/66644
Reviewed By: malfet
Differential Revision: D31662559
Pulled By: msaroufim
fbshipit-source-id: bc8a57b99c2980490ede3974279d1124228a7406
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64382
* This `use_external_data_format` parameter is used for large models cannot be exported because of the 2GB protobuf limit.
* When `use_external_data_format` set to True, the model is exported in ONNX external data format, in which case some of the model parameters are stored in external binary files and not in the ONNX model file itself.
* This PR will set this paramter to DEPRECATED and check the model proto sizes by code instead of by user, if the sizes lager than 2GB, then `use_external_data_format = True` automatically.
Test Plan: Imported from OSS
Reviewed By: ezyang
Differential Revision: D30905265
Pulled By: malfet
fbshipit-source-id: 82b4e17bfa6a8de2bfd700a5282c12f6835603cb
Co-authored-by: hwangdeyu <dejack953@outlook.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64380
* `example_outputs` used to determine the type and shape of the outputs without tracing the execution of the model. And it must be provided when exporting a ScriptModule or ScriptFunction when using export() function.
* Since we can work out `example_outputs` in internal function instead of being provided by user, so we deprecated this argument in the export() function to increase user experience of calling this function.
Test Plan: Imported from OSS
Reviewed By: ezyang
Differential Revision: D30905266
Pulled By: malfet
fbshipit-source-id: d00b00d7d02b365d165028288ad915678caa51f2
Co-authored-by: hwangdeyu <dejack953@outlook.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64373
* Fix some bad formatting and clarify things in onnx.rst.
* In `export_to_pretty_string`:
* Add documentation for previously undocumented args.
* Document that `f` arg is ignored and mark it deprecated.
* Update tests to stop setting `f`.
* Warn if `_retain_param_name` is set.
* Use double quotes for string literals in test_operators.py.
Test Plan: Imported from OSS
Reviewed By: ezyang
Differential Revision: D30905271
Pulled By: malfet
fbshipit-source-id: 3627eeabf40b9516c4a83cfab424ce537b36e4b3
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64371
As of now, the "strip_doc_string" parameter was described as below:
strip_doc_string (bool, default True): do not include the field
doc_string``` from the exported model. Otherwise the field will mention the source code locations for model``.
This is usually useless to users who want to transform a PyTorch model to ONNX one. Only when the user wants to debug the export process, these source code locations could provide benefits.
To make the export() function more friendly by providing less parameters, we combined "strip_doc_string" into "verbose" parameter. If a user set verbose to True, it means the users need some log information for debugging the export process and this is similar with the purpose of strip_doc_string parameter.
But the usage of these 2 arguments are opposite: setting verbose to True means we want to print log information to help debug, which means strip_doc_string should be False. And this is how we replace strip_doc_string with verbose argument in this PR.
This PR will still keep it in torch.onnx.export() function for backward support while the usage of it has been combined with verbose argument.
Test Plan: Imported from OSS
Reviewed By: ezyang
Differential Revision: D30905268
Pulled By: malfet
fbshipit-source-id: 2f06eb805c01fe15ff7a1b4f6595c937ba716d60
Co-authored-by: fatcat-z <zhang-ji@outlook.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64370
As of now, the "_retain_param_name" parameter has no description in PyTorch docs website. According to code, this argument determines if we keep the original parameter names of PyTorch model in the final ONNX graph. If this is False, those original parameter names will be replaced with a series of integers starting from 1.
Since setting numbers as parameter names make no sense to users, we remove this argument from the torch.onnx.export() function to increase user experience of calling this function.
This PR will still keep it in torch.onnx.export() function for backward support while all backend logic has been changed to work as _retain_param_name is set to True.
Test Plan: Imported from OSS
Reviewed By: ezyang
Differential Revision: D30905270
Pulled By: malfet
fbshipit-source-id: ca60757ca17daaff937e9f08da42596086795f4a
Co-authored-by: fatcat-z <zhang-ji@outlook.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64369
As of now, the "enable_onnx_checker" parameter was described as below:
enable_onnx_checker (bool, default True): If True the ONNX model checker will be run to ensure the exported model is a valid ONNX model.
An invalid ONNX graph is useless to users so such checker should be done for each call.
In this PR, we will still write the model to an ONNX file even it is invalid. And the exception will be thrown after the ONNX file has been created. This enables user output an invalid ONNX graph for debug.
This PR will still keep it in torch.onnx.export() function for backward support while all backend logic has been changed to work as enable_onnx_checker is set to True.
Test Plan: Imported from OSS
Reviewed By: ezyang
Differential Revision: D30905267
Pulled By: malfet
fbshipit-source-id: 3ad3f68e77fcec012cc7ef674cc9a61755eebc9e
Co-authored-by: fatcat-z <zhang-ji@outlook.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/62763
This PR is to fix the issue that the graph inputs might be updated when we export the model in inference mode.
When a model is export in inference mode, some optimizations will be made. One side effect of these optimizations is: the inputs of graph might be adjusted. Such optimizatiosn include:
1. Conv and BatchNorm op fusion.
2. Do constant folding.
If the user sets export_params=False, or set keep_initializers_as_inputs=True, it's highly possible that the user wants to provide the corresponding parameters or initiliazers as the inputs of the graph.
In such situation, no matter the model is export in inference mode or training mode, exporter needs to prevent above optimizations from adjusting the graph inputs. By this, the inputs of graph could match inputs that users provided.
The changes in this PR, add an additional common judgement to see if the above optimizations needs to be done or not. From the value of export_params and keep_initializers_as_inputs arguments, infer if the graph inputs are allowed to be adjusted.
If no, these optimizations will be ignored, even other requirements are matched.
Besides these code changes, the comments of some parameters below have been updated so that users have more thoughts when they consider how to leverage these parameters for different purposes:
1. export_params
2. training
3. do_constant_folding
4. keep_initializers_as_inputs
Test Plan: Imported from OSS
Reviewed By: SplitInfinity
Differential Revision: D30375183
Pulled By: msaroufim
fbshipit-source-id: 4db8b9695649eb32a3a0fefa950ee2e5651bdba0
Co-authored-by: fatcat-z <jiz@microsoft.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/60249
* Add introductory paragraph explaining what ONNX is and what the
torch.onnx module does.
* In "Tracing vs Scripting" and doc-string for torch.onnx.export(),
clarify that exporting always happens on ScriptModules and that
tracing and scripting are the two ways to produce a ScriptModule.
* Remove examples of using Caffe2 to run exported models.
Caffe2's website says it's deprecated, so it's probably best not to
encourage people to use it by including it in examples.
* Remove a lot of content that's redundant:
* The example of how to mix tracing and scripting, and instead
link to Introduction to TorchScript, which includes very similar
content.
* "Type annotations" section. Link to TorchScript docs which explain
that in more detail.
* "Using dictionaries to handle Named Arguments as model inputs"
section. It's redundant with the description of the `args` argument
to `export()`, which appears on the same page once the HTML
is generated.
* Remove the list of supported Tensor indexing patterns. If it's not
in the list of unsupported patterns, users can assume it's
supported, so having both is redundant.
* Remove the list of supported operators and models.
I think the list of supported operators is not very useful.
A list of supported model architectures may be useful, but in
reality it's already very out of date. We should add it back if
/ when we have a system for keeping it up to date.
* "Operator Export Type" section. It's redundant with the description
of the `operator_export_type` arg to to `export()`, which appears on
the same page once the HTML is generated.
* "Use external data format" section. It's redundant with the
description of the `use_external_data_format` arg to `export()`.
* "Training" section. It's redundant with the
description of the `training` arg to `export()`.
* Move the content about different operator implementations producing
different results from the "Limitations" section into the doc for the
`operator_export_type` arg.
* Document "quantized" -> "caffe2" behavior of
OperatorExportTypes.ONNX_ATEN_FALLBACK.
* Combing the text about using torch.Tensor.item() and the text about
using NumPy types into a section titled
"Avoid NumPy and built-in Python types", since they're both
fundamentally about the same issue.
* Rename "Write PyTorch model in Torch way" to "Avoiding Pitfalls".
* Lots of minor fixes: spelling, grammar, brevity, fixing links, adding
links.
* Clarify limitation on input and output types. Phrasing it in terms of
PyTorch types is much more accessible than in terms of TorchScript
types. Also clarify what actually happens when dict and str are used
as inputs and outputs.
* In Supported operators, use torch function and class names and link
to them. This is more user friendly than using the internal aten
op names.
* Remove references to VariableType.h, which doesn't appear to contain
the information that it once did. Instead refer to the generated
.pyi files.
* Remove the text in the FAQ about appending to lists within loops.
I think this limitation is no longer present
(perhaps since https://github.com/pytorch/pytorch/pull/51577).
* Minor fixes to some code I read along the way.
* Explain the current rationale for the weird ::prim_PythonOp op name.
Test Plan: Imported from OSS
Reviewed By: zou3519, ZolotukhinM
Differential Revision: D29494912
Pulled By: SplitInfinity
fbshipit-source-id: 7756c010b2320de0692369289604403d28877719
Co-authored-by: Gary Miguel <garymiguel@microsoft.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/58695
As PEP8 says: "Pick a rule and stick to it." [1]
[1] https://www.python.org/dev/peps/pep-0008/#string-quotes
Test Plan: Imported from OSS
Reviewed By: driazati
Differential Revision: D28714811
Pulled By: SplitInfinity
fbshipit-source-id: c95103aceb1725c17c034dc6fc8216627f189548
Co-authored-by: Gary Miguel <garymiguel@microsoft.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/57079
Testing onnx 1.9 release, we see that the old bug is triggered for the caffe2 test:
`pytest test/onnx/test_pytorch_onnx_caffe2_quantized.py::TestQuantizedOps::test_small_model`
This is because the graph inputs
```python
graph(%x.1 : Tensor,
%conv1._packed_params : __torch__.torch.classes.quantized.Conv2dPackedParamsBase,
%conv2._packed_params : __torch__.torch.classes.quantized.Conv2dPackedParamsBase,
%fc.bias : Float(10, strides=[1], requires_grad=0, device=cpu),
%fc.weight : Float(10, 72, strides=[72, 1], requires_grad=0, device=cpu)):
```
contains `Conv2dPackedParamsBase` which is a PackedParams.
When we do flatten, we will flatten to several tensors, then the shape inference for input misaligned.
This PR record how may tensors got flattened in PackeParams, and skip by these number rather than 1, then the UT passed.
Note that tuple case should still follow the original logic.
Test Plan: Imported from OSS
Reviewed By: SplitInfinity
Differential Revision: D28393949
Pulled By: malfet
fbshipit-source-id: 98d48aad27e5ca03fb10d260f8e625478d996ee2
Co-authored-by: David <jiafa@microsoft.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/56172
Enable the standardOps include **Add\Sub\Mul\Div\Gemm\Pow\Mod** with low precision input in ORT
Test Plan: Imported from OSS
Reviewed By: pbelevich
Differential Revision: D27866136
Pulled By: SplitInfinity
fbshipit-source-id: f2cf5649fffefd68c0cc7b6dce94198751636727
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/53307
This PR did symbolic shape inference, in the onnx pass _jit_pass_onnx_graph_shape_type_inference.
It creates a singleton ConstantValueMap.
It leverages constant folding technique and did a per-op based handling for ConstantValueMap.
As a byproduct, it enables fold_if pass for dynamic axes cases, typically for faster-rcnn etc.
The core change is in `torch/csrc/jit/passes/onnx/shape_type_inference.cpp` and `torch/csrc/jit/passes/onnx/constant_map.cpp`.
We usually need copy tensor to store in the ConstantValueMap, otherwise the underlying value may change. I see this issue in (1) from_blob (2) get value from Constant node.
Test Plan: Imported from OSS
Reviewed By: pbelevich, malfet
Differential Revision: D26922414
Pulled By: SplitInfinity
fbshipit-source-id: 7654dc13d1de8d9496ad4be89f1454260d7bdeb0
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/53306
* [ONNX] Fix for sequence of mutations in blocks (#51577)
Fixes consecutive mutations in a tensor inside blocks.
Also, support append and pop in blocks.
* Support inplace operations + indexing
* Clean up old pass for remove mutations
* Add loop test
* Fixes for set attr in loops
* Removing the new jit API flag
* [ONNX] Redesign onnx pass to enable shape type dependent pattern conversion - cont (#51795)
With the introduction of ONNX shape inference, shape and type are inferred on the fly as operators get converted from ATen to ONNX when running symbolic function. This resolves the shape/type requirement for the symbolic functions. The pre-onnx passes however, can not be supported by shape inference, since at that stage the operators in the graph are still ATen operators.
This PR is to update the design of ONNX pass, to enable a mechanism of capturing subgraphs of ATen operators of certain patterns, and convert them later, when shape/type information of upstream operators are available.
The new design will require pre-onnx passes that need shape/type to be written in two parts, encapsulation and conversion.
The encapsulation part will find the nodes of patterns, like how pre-onnx passes were written previously. But instead of converting the nodes, it will encapsulate them into a sub-block of a new placeholder node. This part is called before onnx pass, so it runs before calling symbolic functions.
The conversion part will be called inside the onnx pass. In onnx pass, run_symbolic_func will be called for each node in topological order. When it reaches the placeholder node, the conversion part will be invoked. It will convert the nodes inside the sub-block based on pattern. By that time, it will have shape/type of upstream operators available. After the conversion is complete, the placeholder node will be removed, and nodes inside its sub-block converted. Run_symbolic_func will be called for these nodes, and they will be converted from ATen operator to ONNX operator.
This PR includes several other fixes, listed below.
* ~~replace helper.cpp with onnx_utils.cpp for holding utility functions.~~
* fix EraseNumberTypes on Bool type, the code was outdated that back then Bool type doesn't exist.
* ~~enable onnx shape inference in export with parameter/initializer data.~~
* other code clean ups.
* fix insertion of identity nodes for loop opset 13 sequence output.
~~PR depends on #51603~~
* Fix after merge
* clang
* Fix clang
* Fix clang
* Fix warning message.
* Fixes for non-model param attributes
* Fix for caffe2
* Additional test
* clang
* Skip test for lower opsets
* fix clang-tidy
* Update init.cpp
* Update remove_inplace_ops_for_onnx.cpp
* Update remove_inplace_ops_for_onnx.cpp
* Update remove_inplace_ops_for_onnx.cpp
* Fix for clang formatting
Test Plan: Imported from OSS
Reviewed By: pbelevich, malfet
Differential Revision: D26922416
Pulled By: SplitInfinity
fbshipit-source-id: e7108620b39b6404c594910786c4d275fee59d84
Co-authored-by: Bowen Bao <bowbao@microsoft.com>
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/53304
With the introduction of ONNX shape inference, shape and type are inferred on the fly as operators get converted from ATen to ONNX when running symbolic function. This resolves the shape/type requirement for the symbolic functions. The pre-onnx passes however, can not be supported by shape inference, since at that stage the operators in the graph are still ATen operators.
This PR is to update the design of ONNX pass, to enable a mechanism of capturing subgraphs of ATen operators of certain patterns, and convert them later, when shape/type information of upstream operators are available.
The new design will require pre-onnx passes that need shape/type to be written in two parts, encapsulation and conversion.
The encapsulation part will find the nodes of patterns, like how pre-onnx passes were written previously. But instead of converting the nodes, it will encapsulate them into a sub-block of a new placeholder node. This part is called before onnx pass, so it runs before calling symbolic functions.
The conversion part will be called inside the onnx pass. In onnx pass, run_symbolic_func will be called for each node in topological order. When it reaches the placeholder node, the conversion part will be invoked. It will convert the nodes inside the sub-block based on pattern. By that time, it will have shape/type of upstream operators available. After the conversion is complete, the placeholder node will be removed, and nodes inside its sub-block converted. Run_symbolic_func will be called for these nodes, and they will be converted from ATen operator to ONNX operator.
This PR includes several other fixes, listed below.
* ~~replace helper.cpp with onnx_utils.cpp for holding utility functions.~~
* fix EraseNumberTypes on Bool type, the code was outdated that back then Bool type doesn't exist.
* ~~enable onnx shape inference in export with parameter/initializer data.~~
* other code clean ups.
* fix insertion of identity nodes for loop opset 13 sequence output.
~~PR depends on #51603~~
Test Plan: Imported from OSS
Reviewed By: SplitInfinity
Differential Revision: D26922417
Pulled By: malfet
fbshipit-source-id: 14ed06158d539e2451c2e5e63ba1b32fb0f75095
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/51517
Fix get/set attributes when getting/setting a model parameter.
This PR also fixes inplace ops in If blocks.
Test Plan: Imported from OSS
Reviewed By: pbelevich
Differential Revision: D26203116
Pulled By: SplitInfinity
fbshipit-source-id: bed6ee6dd92b5b43febc8c584a6872290f8fe33f
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/50910
Fixing pytorch/vision#3251 (PR #49410 triggers the torch vision test build failure, on three tests test_faster_rcnn, test_mask_rcnn, test_keypoint_rcnn. )
The offending PR is fine on pytorch UT, because the torchvision and pytorch test has a gap when we merge them - we are using different test API on two sides, therefore causing some discrepancy.
This PR bridge the gap for the above three tests, and disable _jit_pass_onnx_fold_if pass until it gets fixed.
Allow _jit_pass_onnx_fold_if only when dynamic_axes is None.
Test Plan: Imported from OSS
Reviewed By: pbelevich
Differential Revision: D26050886
Pulled By: SplitInfinity
fbshipit-source-id: b765ffe30914261866dcc761f0d0999fd16169e3
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/50905
Adds an additional run of onnx shape inference after constant folding, since initializer may have changed and affected shape inference.
Test Plan: Imported from OSS
Reviewed By: pbelevich
Differential Revision: D26050881
Pulled By: SplitInfinity
fbshipit-source-id: 9e5d69c52b647133cd3a0781988e2ad1d1a9c09d
Summary:
Handle sequence output shape and type inference.
This PR fixes value type of sequence outputs. Prior to this, all model sequence type outputs were unfolded for ONNX models.
This PR also enable shape inference for sequence outputs to represent the dynamic shape of these values.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/46542
Reviewed By: ezyang
Differential Revision: D24924236
Pulled By: bzinodev
fbshipit-source-id: 506e70a38cfe31069191d7f40fc6375239c6aafe
Summary:
The args parameter of ONNX export is changed to better support optional arguments such that args is represented as:
args (tuple of arguments or torch.Tensor, a dictionary consisting of named arguments (optional)):
a dictionary to specify the input to the corresponding named parameter:
- KEY: str, named parameter
- VALUE: corresponding input
Pull Request resolved: https://github.com/pytorch/pytorch/pull/47367
Reviewed By: H-Huang
Differential Revision: D25432691
Pulled By: bzinodev
fbshipit-source-id: 9d4cba73cbf7bef256351f181f9ac5434b77eee8
Summary:
Fixes https://github.com/pytorch/pytorch/issues/45215
Still need to resolve a few mypy issues before a review. In special, there is an error which I don't know how to solve, see:
```python
torch/onnx/utils.py:437: error: Name 'is_originally_training' is not defined [name-defined]
if training is None or training == TrainingMode.EVAL or (training == TrainingMode.PRESERVE and not is_originally_training):
```
`is_originally_training` is used but never defined/imported on [`torch/onnx/utils.py`](ab5cc97fb0/torch/onnx/utils.py (L437)),
Pull Request resolved: https://github.com/pytorch/pytorch/pull/45258
Reviewed By: zhangguanheng66
Differential Revision: D25254920
Pulled By: ezyang
fbshipit-source-id: dc9dc036da43dd56b23bd6141e3ab92e1a16e3b8
Summary:
* Enable ONNX shape inference by default.
* ONNX could potentially set inferred shape in output instead of value_infos, checking both to be sure.
* Small fix in symbol_map to avoid overlooking dup symbols.
* Fix scalar_type_analysis to be consistent with PyTorch scalar type promotion logic.
* Correctly handle None dim_param from ONNX inferred shape.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/46629
Reviewed By: ailzhang
Differential Revision: D24900171
Pulled By: bzinodev
fbshipit-source-id: 83d37fb9daf83a2c5969d8383e4c8aac986c35fb
Summary:
- rand/randn: the type signature of int[] is different in scripting, thus failing the check.
- where: scripting produces dynamic cases which are supported by `unbind` export of higher opsets.
- test_list_pass: this test fails when using new scripting api, should be fixed by https://github.com/pytorch/pytorch/issues/45369
Pull Request resolved: https://github.com/pytorch/pytorch/pull/45793
Reviewed By: mrshenli
Differential Revision: D24566096
Pulled By: bzinodev
fbshipit-source-id: 6fe0925c66dee342106d71c9cbc3c95cabe639f7
Summary:
* Support propagating `dim_param` in ONNX by encoding as `ShapeSymbol` in `SymbolicShape` of outputs. If export is called with `dynamic_axes` provided, shape inference will start with these axes set as dynamic.
* Add new test file `test_pytorch_onnx_shape_inference.py`, reusing all test cases from `test_pytorch_onnx_onnxruntime.py`, but focus on validating shape for all nodes in graph. Currently this is not enabled in the CI, since there are still quite some existing issues and corner cases to fix. The test is default to run only at opset 12.
* Bug fixes, such as div, _len, and peephole.cpp passes for PackPadded, and LogSoftmaxCrossEntropy.
* This PR depends on existing PR such as 44332.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/44920
Reviewed By: eellison
Differential Revision: D23958398
Pulled By: bzinodev
fbshipit-source-id: 00479d9bd19c867d526769a15ba97ec16d56e51d
Summary:
Export of embedding bag with dynamic list of offsets.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/44693
Reviewed By: malfet
Differential Revision: D23831980
Pulled By: bzinodev
fbshipit-source-id: 3eaff1a0f20d1bcfb8039e518d78c491be381e1a
Summary:
Moved description of tool and changes in function name
Pull Request resolved: https://github.com/pytorch/pytorch/pull/44124
Reviewed By: albanD
Differential Revision: D23674618
Pulled By: bzinodev
fbshipit-source-id: 5db0bb14fc106fc96358b1e0590f08e975388c6d
Summary:
* Support sequence type (de)serialization, enables onnx shape inference on sequence nodes.
* Fix shape inference with block input/output: e.g. Loop and If nodes.
* Fix bugs in symbolic discovered by coverage of onnx shape inference.
* Improve debuggability: added more jit logs. For simplicity, the default log level, when jit log is enabled, will not dump ir graphs.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/43929
Reviewed By: albanD
Differential Revision: D23674604
Pulled By: bzinodev
fbshipit-source-id: ab6aacb16d0e3b9a4708845bce27c6d65e567ba7
Summary:
Follow up to https://github.com/pytorch/pytorch/pull/36404
Adding prim::device and prim::dtype to list of skipped peepholes when we run inlining. In the long term another fix may not be to encode shape / dtype info on the traced graph, because it is not guaranteed to be correct. This is blocked by ONNX currently.
Partial fix for https://github.com/pytorch/pytorch/issues/43134
Pull Request resolved: https://github.com/pytorch/pytorch/pull/43363
Reviewed By: glaringlee
Differential Revision: D23383987
Pulled By: eellison
fbshipit-source-id: 2e9c5160d39d690046bd9904be979d58af8d3a20
Summary:
Duplicate of https://github.com/pytorch/pytorch/issues/41413
This PR initiates the process of updating the torchsciprt backend interface used by ONNX exporter.
Replace jit lower graph pass by freeze module pass
Enable ScriptModule tests for ONNX operator tests (ORT backend) and model tests by default.
Replace jit remove_inplace_ops pass with remove_mutation and consolidation all passes for handling inplace ops.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/43791
Reviewed By: houseroad
Differential Revision: D23421872
Pulled By: bzinodev
fbshipit-source-id: a98710c45ee905748ec58385e2a232de2486331b
Summary:
It is often that the conversion from torch operator to onnx operator requires input rank/dtype/shape to be known. Previously, the conversion depends on tracer to provide these info, leaving a gap in conversion of scripted modules.
We are extending the export with support from onnx shape inference. If enabled, onnx shape inference will be called whenever an onnx node is created. This is the first PR introducing the initial look of the feature. More and more cases will be supported following this PR.
* Added pass to run onnx shape inference on a given node. The node has to have namespace `onnx`.
* Moved helper functions from `export.cpp` to a common place for re-use.
* This feature is currently experimental, and can be turned on through flag `onnx_shape_inference` in internal api `torch.onnx._export`.
* Currently skipping ONNX Sequence ops, If/Loop and ConstantOfShape due to limitations. Support will be added in the future.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/40628
Reviewed By: mrshenli
Differential Revision: D22709746
Pulled By: bzinodev
fbshipit-source-id: b52aeeae00667e66e0b0c1144022f7af9a8b2948
Summary:
The ONNX spec for the Squeeze operator:
> Remove single-dimensional entries from the shape of a tensor. Takes a parameter axes with a list of axes to squeeze. If axes is not provided, all the single dimensions will be removed from the shape. If an axis is selected with shape entry not equal to one, an error is raised.
Currently, as explained in issue https://github.com/pytorch/pytorch/issues/36796, it is possible to export such a model to ONNX, and this results in an exception from ONNX runtime.
Fixes https://github.com/pytorch/pytorch/issues/36796.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/38476
Reviewed By: hl475
Differential Revision: D22158024
Pulled By: houseroad
fbshipit-source-id: bed625f3c626eabcbfb2ea83ec2f992963defa19
Summary:
in `_jit_pass_onnx`, symbolic functions are called for each node for conversion. However, there are nodes that cannot be converted without additional context. For example, the number of outputs from split (and whether it is static or dynamic) is unknown until the point where it is unpacked by listUnpack node. This pass does a preprocess, and prepares the nodes such that enough context can be received by the symbolic function.
* After preprocessing, `_jit_pass_onnx` should have enough context to produce valid ONNX nodes, instead of half baked nodes that replies on fixes from later postpasses.
* `_jit_pass_onnx_peephole` should be a pass that does ONNX specific optimizations instead of ONNX specific fixes.
* Producing more valid ONNX nodes in `_jit_pass_onnx` enables better utilization of the ONNX shape inference https://github.com/pytorch/pytorch/issues/40628.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/41832
Reviewed By: ZolotukhinM
Differential Revision: D22968334
Pulled By: bzinodev
fbshipit-source-id: 8226f03c5b29968e8197d242ca8e620c6e1d42a5
Summary:
* move both under new file `fixup_onnx_controlflow`
* move the fixup to where the ONNX loop/if node is created, as oppose to running the fixup as postpass. This will help with enable onnx shape inference later.
* move `fuseSequenceSplitConcat` to `Peephole`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/40943
Reviewed By: mrshenli
Differential Revision: D22709999
Pulled By: bzinodev
fbshipit-source-id: 51d316991d25dc4bb4047a6bb46ad1e2401d3d2d
Summary:
This PR initiates the process of updating the torchsciprt backend interface used by ONNX exporter.
- Replace jit lower graph pass by freeze module pass
- Enable ScriptModule tests for ONNX operator tests (ORT backend) and model tests by default.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/41413
Reviewed By: VitalyFedyunin
Differential Revision: D22845258
Pulled By: bzinodev
fbshipit-source-id: d57fd4086f27bd0c3bf5f70af7fd0daa39a2814a
