IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[DOCS] PORT Add examples of using named outputs in extensions (#17622) (#17754)

Browse Source 
* [DOCS] Add examples of using named outputs in extensions (#17622)

* [DOCS]Add examples of using named outputs in extensions

* Fix opset

* Apply suggestions from code review

Co-authored-by: Tatiana Savina <tatiana.savina@intel.com>

* Update docs/Extensibility_UG/frontend_extensions.md

* Add reference to external docs

* Update docs/Extensibility_UG/frontend_extensions.md

Co-authored-by: Tatiana Savina <tatiana.savina@intel.com>

---------

Co-authored-by: Tatiana Savina <tatiana.savina@intel.com>

* Update docs/Extensibility_UG/frontend_extensions.md

---------

Co-authored-by: Maxim Vafin <maxim.vafin@intel.com>
Co-authored-by: Karol Blaszczak <karol.blaszczak@intel.com>
This commit is contained in:
Tatiana Savina 2023-06-15 16:35:45 +02:00 committed by GitHub
parent 9684f9184a
commit 8031cfea98
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 62 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
docs/Extensibility_UG/frontend_extensions.md
View File

@ -381,6 +381,25 @@ To access original framework operation attribute value and connect to inputs, ``
The conversion function should return a vector of node outputs that are mapped to
corresponding outputs of the original framework operation in the same order.
Some frameworks require output names of the operation to be provided during conversion.
For PaddlePaddle operations, it is generally necessary to provide names for all outputs using the ``NamedOutputs`` container.
Usually those names can be found in source code of the individual operation in PaddlePaddle code.
The following example shows such conversion for the ``top_k_v2`` operation.
.. doxygensnippet:: docs/snippets/ov_extensions.cpp
:language: cpp
:fragment: [frontend_extension_paddle_TopK]
For TensorFlow framework, if an operation has more than one output, it is recommended to assign names to
those outputs using the ``NamedOutputVector`` structure which allows both indexed and named output access.
For a description of TensorFlow operations, including the names of their outputs, refer to the
`tf.raw_ops <https://www.tensorflow.org/api_docs/python/tf/raw_ops/>`__ documentation page.
The next example shows such conversion for the ``TopKV2`` operation.
.. doxygensnippet:: docs/snippets/ov_extensions.cpp
:language: cpp
:fragment: [frontend_extension_tf_TopK]
@endsphinxdirective

48
docs/snippets/ov_extensions.cpp
View File

@ -2,6 +2,8 @@
// SPDX-License-Identifier: Apache-2.0
//
#include <openvino/openvino.hpp>
#include <openvino/frontend/node_context.hpp>
#include <openvino/frontend/paddle/node_context.hpp>
//! [add_extension_header]
//#include <openvino/core/op_extension.hpp>
//! [add_extension_header]
@ -14,7 +16,7 @@
//! [frontend_extension_Identity_header]
//! [frontend_extension_ThresholdedReLU_header]
#include <openvino/opsets/opset8.hpp>
#include <openvino/opsets/opset11.hpp>
//! [frontend_extension_ThresholdedReLU_header]
//! [frontend_extension_framework_map_macro_headers]
@ -204,13 +206,49 @@ auto extension = std::make_shared<ov::frontend::OpExtension<CustomElu>>("aten::e
core.add_extension(ov::frontend::ConversionExtension(
"ThresholdedRelu",
[](const ov::frontend::NodeContext& node) {
auto greater = std::make_shared<ov::opset8::Greater>(
auto greater = std::make_shared<ov::opset11::Greater>(
node.get_input(0),
ov::opset8::Constant::create(ov::element::f32, {}, {node.get_attribute<float>("alpha")}));
auto casted = std::make_shared<ov::opset8::Convert>(greater, ov::element::f32);
return ov::OutputVector{ std::make_shared<ov::opset8::Multiply>(node.get_input(0), casted) };
ov::opset11::Constant::create(ov::element::f32, {}, {node.get_attribute<float>("alpha")}));
auto casted = std::make_shared<ov::opset11::Convert>(greater, ov::element::f32);
return ov::OutputVector{ std::make_shared<ov::opset11::Multiply>(node.get_input(0), casted) };
}));
//! [frontend_extension_ThresholdedReLU]
//! [frontend_extension_paddle_TopK]
core.add_extension(ov::frontend::ConversionExtension("top_k_v2", [](const ov::frontend::NodeContext& node) {
auto x = node.get_input("X");
const auto k_expected = node.get_attribute<int>("k", 1);
auto k_expected_node = ov::opset11::Constant::create(ov::element::i32, {}, {k_expected});
auto axis = node.get_attribute<int32_t>("axis", -1);
bool sorted = node.get_attribute<bool>("sorted", true);
bool largest = node.get_attribute<bool>("largest", true);
std::string sort_type = sorted ? "value" : "none";
std::string mode = largest ? "max" : "min";
auto node_topk = std::make_shared<ov::opset11::TopK>(x, k_expected_node, axis, mode, sort_type);
ov::frontend::paddle::NamedOutputs named_outputs;
named_outputs["Out"] = ov::OutputVector{node_topk->output(0)};
named_outputs["Indices"] = ov::OutputVector{node_topk->output(1)};
return named_outputs;
}));
//! [frontend_extension_paddle_TopK]
//! [frontend_extension_tf_TopK]
core.add_extension(ov::frontend::ConversionExtension("TopKV2", [](const ov::frontend::NodeContext& node) {
auto input = node.get_input(0);
auto k_input = node.get_input(1);
bool sorted = node.get_attribute<bool>("sorted", true);
auto mode = ov::opset11::TopK::Mode::MAX;
auto sort_type = sorted ? ov::opset11::TopK::SortType::SORT_VALUES : ov::opset11::TopK::SortType::SORT_INDICES;
auto top_k = std::make_shared<ov::opset11::TopK>(input, k_input, -1, mode, sort_type, ov::element::i32, true);
return ov::frontend::NamedOutputVector{{"values", top_k->output(0)}, {"indices", top_k->output(1)}};
}));
//! [frontend_extension_tf_TopK]
}
{
//! [add_extension_lib]