Port api reference (#11152)

Author: ilyachur

docs/IE_PLUGIN_DG/Intro.md

@@ -9,11 +9,12 @@
 
    Implement Plugin Functionality <openvino_docs_ie_plugin_dg_plugin>
    Implement Executable Network Functionality <openvino_docs_ie_plugin_dg_executable_network>
-   openvino_docs_ie_plugin_dg_quantized_networks
    Implement Synchronous Inference Request <openvino_docs_ie_plugin_dg_infer_request>
    Implement Asynchronous Inference Request <openvino_docs_ie_plugin_dg_async_infer_request>
    openvino_docs_ie_plugin_dg_plugin_build
    openvino_docs_ie_plugin_dg_plugin_testing
+   openvino_docs_ie_plugin_detailed_guides
+   openvino_docs_ie_plugin_api_references
 
 @endsphinxdirective
 
@@ -61,5 +62,5 @@ Detailed guides
 API References
 -----------------------
 
-* [Inference Engine Plugin API](groupie_dev_api.html)
-* [Inference Engine Transformation API](groupie_transformation_api.html)
+* [Inference Engine Plugin API](@ref ie_dev_api)
+* [Inference Engine Transformation API](@ref ie_transformation_api)

docs/IE_PLUGIN_DG/detailed_guides.md

@@ -0,0 +1,18 @@
+# Advanced Topics {#openvino_docs_ie_plugin_detailed_guides}
+
+@sphinxdirective
+
+.. toctree::
+   :maxdepth: 1
+   :hidden:
+
+   openvino_docs_ie_plugin_dg_quantized_networks
+   openvino_docs_IE_DG_lpt
+
+@endsphinxdirective
+
+The guides below provides extra information about specific features of OpenVINO needed for understanding during OpenVINO plugin development:
+
+* [Quantized networks](@ref openvino_docs_ie_plugin_dg_quantized_networks)
+* [Low precision transformations](@ref openvino_docs_IE_DG_lpt) guide
+* [Writing OpenVINO™ transformations](@ref openvino_docs_transformations) guide

docs/IE_PLUGIN_DG/dev_api_references.md

@@ -0,0 +1,17 @@
+# Plugin API Reference {#openvino_docs_ie_plugin_api_references}
+
+@sphinxdirective
+
+.. toctree::
+   :maxdepth: 1
+   :hidden:
+
+   ../groupie_dev_api
+   ../groupie_transformation_api
+
+@endsphinxdirective
+
+The guides below provides extra API references needed for OpenVINO plugin development:
+
+* [OpenVINO Plugin API](@ref ie_dev_api)
+* [OpenVINO Transformation API](@ref ie_transformation_api)

docs/IE_PLUGIN_DG/plugin_transformation_pipeline/PluginTransformationPipeline.md

@@ -10,6 +10,7 @@ API references available:
 
 .. toctree::
    :maxdepth: 2
-   
-   ../groupie_cpp_api
+
+   ../groupov_cpp_api
+   ../groupie_c_api
    ie_python_api/api

docs/api/api_reference.rst

@@ -72,9 +72,7 @@
 
    openvino_docs_Extensibility_UG_Intro
    openvino_docs_transformations
-   Inference Engine Plugin Developer Guide <openvino_docs_ie_plugin_dg_overview>
-   groupie_dev_api
-   Plugin Transformation Pipeline <openvino_docs_IE_DG_plugin_transformation_pipeline>
+   OpenVINO Plugin Developer Guide <openvino_docs_ie_plugin_dg_overview>
 
 .. toctree::
    :maxdepth: 1

docs/documentation.md

@@ -79,7 +79,7 @@ EXTRA_PAGE_LIST = {}
 --! is not set (otherwise, the title of intro file will be used).
 --!
 
-INDEX_TITLE = "Inference Engine C++ API Reference"
+INDEX_TITLE = "OpenVINO Runtime C++ API Reference"
 
 --!
 --! File with project introduction (reStructuredText). When non-nil, this file

docs/doxyrest-config.lua

@@ -12,7 +12,6 @@
  */
 
 /**
- * @ingroup ie_cpp_api
  * @defgroup ie_transformation_api Inference Engine Transformation API
  * @brief Defines Inference Engine Transformations API which is used to transform ngraph::Function
  *

src/common/transformations/include/transformations_visibility.hpp

@@ -11,6 +11,35 @@
 // OPENVINO_API is used for the public API symbols. It either DLL imports or DLL exports
 //    (or does nothing for static build)
 
+/**
+ * @defgroup ov_cpp_api OpenVINO Runtime C++ API
+ * OpenVINO Runtime C++ API
+ *
+ * @defgroup ov_model_cpp_api Basics
+ * @ingroup ov_cpp_api
+ * OpenVINO Core C++ API to work with ov::Model, dynamic and static shapes, types
+ *
+ * @defgroup ov_ops_cpp_api Operations
+ * @ingroup ov_cpp_api
+ * OpenVINO C++ API to create operations from different opsets. Such API is used to
+ * creation models from code, write transformations and traverse the model graph
+ *
+ * @defgroup ov_opset_cpp_api Operation sets
+ * @ingroup ov_cpp_api
+ * OpenVINO C++ API to work with operation sets
+ *
+ * @defgroup ov_runtime_cpp_api Inference
+ * @ingroup ov_cpp_api
+ * OpenVINO Inference C++ API provides ov::Core, ov::CompiledModel, ov::InferRequest
+ * and ov::Tensor classes
+ */
+
+/**
+ * @brief OpenVINO C++ API
+ * @ingroup ov_cpp_api
+ */
+namespace ov {}  // namespace ov
+
 #ifdef _WIN32
 #    pragma warning(disable : 4251)
 #    pragma warning(disable : 4275)

src/core/include/openvino/core/core_visibility.hpp

@@ -34,7 +34,11 @@ class FrontEnd;
 }
 
 class ModelAccessor;
-/// A user-defined model.
+
+/**
+ * @brief A user-defined model
+ * @ingroup ov_model_cpp_api
+ */
 class OPENVINO_API Model : public std::enable_shared_from_this<Model> {
     friend class frontend::FrontEnd;
     friend OPENVINO_API std::shared_ptr<Model> clone_model(const Model& func,

src/core/include/openvino/core/model.hpp

@@ -61,6 +61,7 @@ class Tensor;
 /**
  * @brief Wraps allocator implementation to provide safe way to store allocater loaded from shared library
  *        And constructs default based on `new` `delete` c++ calls allocator if created without parameters
+ * @ingroup ov_runtime_cpp_api
  */
 class OPENVINO_API Allocator {
     AllocatorImpl::Ptr _impl;

src/core/include/openvino/runtime/allocator.hpp

@@ -30,8 +30,8 @@ class VariableState;
 
 /**
  * @brief Tensor API holding host memory
- *
  * It can throw exceptions safely for the application, where it is properly handled.
+ * @ingroup ov_runtime_cpp_api
  */
 class OPENVINO_API Tensor {
 protected:
@@ -208,6 +208,9 @@ class OPENVINO_API Tensor {
     }
 };
 
+/**
+ * @brief A vector of Tensor's
+ */
 using TensorVector = std::vector<Tensor>;
 
 namespace runtime {

src/core/include/openvino/runtime/tensor.hpp

@@ -2,16 +2,11 @@
 // SPDX-License-Identifier: Apache-2.0
 //
 
-/**
- *  @defgroup ie_cpp_api Inference Engine C++ API
- *  Inference Engine C++ API
- */
-
 /**
  * @brief The macro defines a symbol import/export mechanism essential for Microsoft Windows(R) OS.
- *
  * @file ie_api.h
  */
+
 #pragma once
 
 #if defined(OPENVINO_STATIC_LIBRARY) || defined(USE_STATIC_IE) || (defined(__GNUC__) && (__GNUC__ < 4))

src/inference/include/ie/ie_api.h

@@ -27,7 +27,6 @@
 #include "ie_api.h"
 
 /**
- * @ingroup ie_cpp_api
  * @brief Inference Engine C++ API
  */
 namespace InferenceEngine {

src/inference/include/ie/ie_version.hpp

@@ -46,6 +46,7 @@ namespace ie = InferenceEngine;
 
 /**
  * @brief This type of map is used for result of Core::query_model
+ * @ingroup ov_runtime_cpp_api
  *   - `key` means operation name
  *   - `value` means device name supporting this operation
  */

src/inference/include/openvino/runtime/common.hpp

@@ -32,6 +32,7 @@ class InferRequest;
 
 /**
  * @brief This class represents a compiled model.
+ * @ingroup ov_runtime_cpp_api
  * A model is compiled by a specific device by applying multiple optimization
  * transformations, then mapping to compute kernels.
  */

src/inference/include/openvino/runtime/compiled_model.hpp

@@ -34,6 +34,7 @@ namespace ov {
 
 /**
  * @brief This class represents an OpenVINO runtime Core entity.
+ * @ingroup ov_runtime_cpp_api
  * User applications can create several Core class instances, but in this case the underlying plugins
  * are created multiple times and not shared between several Core instances. The recommended way is to have
  * a single Core instance per application.

src/inference/include/openvino/runtime/core.hpp

@@ -11,16 +11,19 @@ namespace ov {
 
 /**
  * @brief Thrown in case of cancelled asynchronous operation.
+ * @ingroup ov_runtime_cpp_api
  */
 class OPENVINO_RUNTIME_API Cancelled : public Exception {
     using Exception::Exception;
 };
 
 /**
- * @brief Thrown in case of calling the InferRequest methods while the request is busy with compute operation.
+ * @brief Thrown in case of calling the InferRequest methods while the request is
+ * busy with compute operation.
+ * @ingroup ov_runtime_cpp_api
  */
 class OPENVINO_RUNTIME_API Busy : public Exception {
     using Exception::Exception;
 };
 
-}  // namespace ov
+}  // namespace ov

src/inference/include/openvino/runtime/exception.hpp

@@ -29,6 +29,7 @@ class CompiledModel;
 
 /**
  * @brief This is a class of infer request that can be run in asynchronous or synchronous manners.
+ * @ingroup ov_runtime_cpp_api
  */
 class OPENVINO_RUNTIME_API InferRequest {
     std::shared_ptr<InferenceEngine::IInferRequestInternal> _impl;

src/inference/include/openvino/runtime/infer_request.hpp

@@ -14,6 +14,12 @@
 
 namespace ov {
 
+/**
+ * @defgroup ov_runtime_gna_prop_cpp_api Intel GNA specific properties
+ * @ingroup ov_runtime_cpp_api
+ * Set of Intel GNA specific properties.
+ */
+
 /**
  * @brief Namespace with Intel GNA specific properties
  */
@@ -22,12 +28,14 @@ namespace intel_gna {
 /**
  * @brief Property to get an std::string of GNA Library version, usually in the form
  * <API_REVISION>.<RELEASE_LINE>.<RELEASE>.<BUILD>
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<std::string, PropertyMutability::RO> library_full_version{"GNA_LIBRARY_FULL_VERSION"};
 
 /**
  * @brief Scale factor provided by the user to use static quantization.
  * This option should be used with floating point value serialized to string with . (dot) as a decimal separator
+ * @ingroup ov_runtime_gna_prop_cpp_api
  * @details In the case of multiple inputs, individual scale factors can be provided using the
  *  map where key is layer name and value is scale factor
  * Example:
@@ -45,11 +53,13 @@ static constexpr Property<std::map<std::string, float>> scale_factors_per_input{
 
 /**
  * @brief if turned on, dump GNA firmware model into specified file
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<std::string> firmware_model_image_path{"GNA_FIRMWARE_MODEL_IMAGE"};
 
 /**
  * @brief Enum to define software acceleration mode
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 enum class ExecutionMode {
     AUTO = 0,  //!< Uses Intel GNA if available, otherwise uses software execution mode on CPU.
@@ -103,6 +113,7 @@ inline std::istream& operator>>(std::istream& is, ExecutionMode& execution_mode)
 
 /**
  * @brief Enum to define HW compile and execution targets
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 enum class HWGeneration {
     UNDEFINED = 0,  //!< GNA HW generation is undefined
@@ -143,6 +154,7 @@ inline std::istream& operator>>(std::istream& is, HWGeneration& hw_generation) {
 /**
  * @brief GNA proc_type setting that should be one of AUTO, HW, GNA_HW_WITH_SW_FBACK,
  *        GNA_SW_EXACT or SW_FP32
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<ExecutionMode> execution_mode{"GNA_DEVICE_MODE"};
 
@@ -153,22 +165,26 @@ static constexpr Property<ExecutionMode> execution_mode{"GNA_DEVICE_MODE"};
  * If HW is not present, use the option corresponding to the latest fully supported GNA HW generation.
  * A fully supported GNA HW generation means it must be supported by both the OV GNA Plugin and the core GNA Library.
  * Currently, the latest supported GNA HW generation corresponds to GNA_3_0.
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<HWGeneration> execution_target{"GNA_HW_EXECUTION_TARGET"};
 
 /**
  * @brief The option to override the GNA HW compile target. May be one of GNA_2_0, GNA_3_0.
  * By default the same as execution_target.
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<HWGeneration> compile_target{"GNA_HW_COMPILE_TARGET"};
 
 /**
  * @brief if enabled produced minimum memory footprint for compiled model in GNA memory, default value is true
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<bool> memory_reuse{"GNA_COMPACT_MODE"};
 
 /**
  * @brief Enum to define PWL design algorithm
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 enum class PWLDesignAlgorithm {
     UNDEFINED = 0,             //!< PWL approximation algorithm is undefined
@@ -213,13 +229,15 @@ inline std::istream& operator>>(std::istream& is, PWLDesignAlgorithm& pwl_design
  * If value is UNIFORM_DISTRIBUTION then simple uniform distribution is used to create
  * PWL approximation of activation functions.
  * Uniform distribution usually gives poor approximation with the same number of segments
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<PWLDesignAlgorithm> pwl_design_algorithm{"GNA_PWL_DESIGN_ALGORITHM"};
 
 /**
  * @brief The option to allow to specify the maximum error percent that the optimized algorithm finding
  * will be used to find PWL functions.
  * By default (in case of NO value set), 1.0 value is used.
+ * @ingroup ov_runtime_gna_prop_cpp_api
  */
 static constexpr Property<float> pwl_max_error_percent{"GNA_PWL_MAX_ERROR_PERCENT"};