[CPU] [TRANSFORMATIONS] FakeConvert decomposition transformation (#28118

)

### Details:
 - *Implement FakeConvert decomposition transformation.*

### Tickets:
 - *[CVS-156963](https://jira.devtools.intel.com/browse/CVS-156963)*

### Prerequisites:
- *#27949

src/common/transformations/include/transformations/op_conversions/fake_convert_decomposition.hpp

@@ -0,0 +1,32 @@
+// Copyright (C) 2024 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#pragma once
+
+#include "openvino/pass/matcher_pass.hpp"
+#include "transformations_visibility.hpp"
+
+namespace ov {
+namespace pass {
+
+class TRANSFORMATIONS_API FakeConvertDecomposition;
+
+}  // namespace pass
+}  // namespace ov
+
+/**
+ * @ingroup ov_transformation_common_api
+ * @brief FakeConvertDecomposition transformation decomposes FakeConvert layer.
+ * f8: f8e4m3, f8e5m2
+ * downconvert: f32->f8, f16->f8, bf16->f8
+ * upconvert: f8->f32, f8->f16, f8->bf16
+ * output = (upconvert(downconvert(input * scale - shift)) + shift) / scale
+ *
+ */
+
+class ov::pass::FakeConvertDecomposition : public ov::pass::MatcherPass {
+public:
+    OPENVINO_MATCHER_PASS_RTTI("FakeConvertDecomposition");
+    FakeConvertDecomposition();
+};

src/common/transformations/src/transformations/op_conversions/fake_convert_decomposition.cpp

@@ -0,0 +1,76 @@
+// Copyright (C) 2024 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#include "transformations/op_conversions/fake_convert_decomposition.hpp"
+
+#include "itt.hpp"
+#include "openvino/core/rt_info.hpp"
+#include "openvino/op/add.hpp"
+#include "openvino/op/constant.hpp"
+#include "openvino/op/convert.hpp"
+#include "openvino/op/divide.hpp"
+#include "openvino/op/fake_convert.hpp"
+#include "openvino/op/multiply.hpp"
+#include "openvino/op/subtract.hpp"
+#include "openvino/pass/pattern/op/wrap_type.hpp"
+
+ov::pass::FakeConvertDecomposition::FakeConvertDecomposition() {
+    MATCHER_SCOPE(FakeConvertDecomposition);
+    auto data = pattern::any_input();
+
+    auto fake_convert = ov::pass::pattern::wrap_type<ov::op::v13::FakeConvert>();
+
+    matcher_pass_callback callback = [OV_CAPTURE_CPY_AND_THIS](ov::pass::pattern::Matcher& m) {
+        auto& pattern_to_output = m.get_pattern_value_map();
+        const auto fake_convert_node =
+            ov::as_type_ptr<ov::op::v13::FakeConvert>(pattern_to_output.at(fake_convert).get_node_shared_ptr());
+
+        if (fake_convert_node == nullptr || transformation_callback(fake_convert_node)) {
+            return false;
+        }
+
+        Output<Node> data{fake_convert_node->input_value(0)};
+        const Output<Node> input_scale{fake_convert_node->input_value(1)};
+        auto input_type = data.get_element_type();
+
+        ov::pass::NodeRegistry decomp_ops;
+        if (input_type != input_scale.get_element_type()) {
+            input_type = input_scale.get_element_type();
+            data = std::make_shared<ov::op::v0::Convert>(data, input_type);
+            data = decomp_ops.add(data.get_node_shared_ptr());
+        }
+
+        std::shared_ptr<Node> result;
+        const auto scale = decomp_ops.make<ov::op::v1::Multiply>(data, input_scale);
+        if (fake_convert_node->get_input_size() == 2) {
+            const auto downconvert =
+                decomp_ops.make<ov::op::v0::Convert>(scale, fake_convert_node->get_destination_element_type());
+            const auto upconvert = decomp_ops.make<ov::op::v0::Convert>(downconvert, input_type);
+
+            result = decomp_ops.make<ov::op::v1::Divide>(upconvert, input_scale);
+        } else {
+            const Output<Node> input_shift{fake_convert_node->input_value(2)};
+            const auto shift = decomp_ops.make<ov::op::v1::Subtract>(scale, input_shift);
+
+            const auto downconvert =
+                decomp_ops.make<ov::op::v0::Convert>(shift, fake_convert_node->get_destination_element_type());
+            const auto upconvert = decomp_ops.make<ov::op::v0::Convert>(downconvert, input_type);
+
+            const auto deshift = decomp_ops.make<ov::op::v1::Add>(upconvert, input_shift);
+            result = decomp_ops.make<ov::op::v1::Divide>(deshift, input_scale);
+        }
+
+        if (result->get_output_element_type(0) != fake_convert_node->get_output_element_type(0)) {
+            result = decomp_ops.make<ov::op::v0::Convert>(result, fake_convert_node->get_output_element_type(0));
+        }
+
+        result->set_friendly_name(m.get_match_root()->get_friendly_name());
+        ov::copy_runtime_info(fake_convert_node, decomp_ops.get());
+        ov::replace_node(m.get_match_root(), result);
+        return true;
+    };
+
+    auto m = std::make_shared<ov::pass::pattern::Matcher>(fake_convert, matcher_name);
+    register_matcher(m, callback);
+}

src/common/transformations/tests/op_conversions/fake_convert_decomposition_test.cpp

@@ -0,0 +1,149 @@
+// Copyright (C) 2024 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#include "transformations/op_conversions/fake_convert_decomposition.hpp"
+
+#include <gtest/gtest.h>
+
+#include "common_test_utils/common_utils.hpp"
+#include "common_test_utils/ov_test_utils.hpp"
+#include "openvino/opsets/opset1.hpp"
+#include "openvino/opsets/opset13.hpp"
+
+using namespace ov;
+
+using FakeConvertDecompositionParams = std::tuple<Shape,            // data shape
+                                                  Shape,            // scale shape
+                                                  Shape,            // shift shape
+                                                  element::Type_t,  // input precision
+                                                  element::Type_t,  // destination precision
+                                                  bool>;            // default shift
+
+class FakeConvertDecompositionTest : public ov::test::TestsCommon,
+                                     public ::testing::WithParamInterface<FakeConvertDecompositionParams> {
+public:
+    static std::string getTestCaseName(::testing::TestParamInfo<FakeConvertDecompositionParams> obj) {
+        FakeConvertDecompositionParams params = obj.param;
+
+        Shape data_shape, scale_shape, shift_shape;
+        element::Type_t data_prec, dst_prec;
+        bool default_shift;
+        std::tie(data_shape, scale_shape, shift_shape, data_prec, dst_prec, default_shift) = params;
+
+        std::ostringstream result;
+        result << "dataShape=" << ov::test::utils::vec2str(data_shape) << "_";
+        result << "scaleShape=" << ov::test::utils::vec2str(scale_shape) << "_";
+        result << "shiftShape=" << ov::test::utils::vec2str(shift_shape) << "_";
+        result << "dataPrecision=" << element::Type(data_prec) << "_";
+        result << "destinationPrecision=" << element::Type(dst_prec) << "_";
+        if (default_shift)
+            result << "defaultShift=true";
+        else
+            result << "defaultShift=false";
+        return result.str();
+    }
+};
+
+TEST_P(FakeConvertDecompositionTest, CompareFunctions) {
+    FakeConvertDecompositionParams params = this->GetParam();
+
+    Shape data_shape, scale_shape, shift_shape;
+    element::Type_t data_prec, dst_prec;
+    bool default_shift;
+    std::tie(data_shape, scale_shape, shift_shape, data_prec, dst_prec, default_shift) = params;
+
+    std::shared_ptr<ov::Model> model(nullptr);
+    {
+        const auto data = std::make_shared<opset1::Parameter>(data_prec, PartialShape(data_shape));
+        const auto scale = std::make_shared<opset1::Constant>(data_prec, scale_shape);
+        const auto shift = std::make_shared<opset1::Constant>(data_prec, shift_shape);
+
+        const auto fake_convert = default_shift ? std::make_shared<opset13::FakeConvert>(data, scale, dst_prec)
+                                                : std::make_shared<opset13::FakeConvert>(data, scale, shift, dst_prec);
+        model = std::make_shared<ov::Model>(NodeVector{fake_convert}, ParameterVector{data});
+
+        pass::Manager manager;
+        manager.register_pass<ov::pass::InitNodeInfo>();
+        manager.register_pass<ov::pass::FakeConvertDecomposition>();
+        manager.run_passes(model);
+
+        OV_ASSERT_NO_THROW(check_rt_info(model));
+    }
+
+    std::shared_ptr<ov::Model> model_ref(nullptr);
+    {
+        const auto input_data = std::make_shared<opset1::Parameter>(data_prec, PartialShape(data_shape));
+        const auto input_scale = std::make_shared<opset1::Constant>(data_prec, scale_shape);
+        const auto input_shift = std::make_shared<opset1::Constant>(data_prec, shift_shape);
+        ParameterVector params;
+        params.push_back(input_data);
+        std::shared_ptr<Node> data = input_data;
+
+        std::shared_ptr<Node> result;
+        const auto scale = std::make_shared<ov::op::v1::Multiply>(data, input_scale);
+        if (default_shift) {
+            const auto downconvert = std::make_shared<ov::op::v0::Convert>(scale, dst_prec);
+            const auto upconvert = std::make_shared<ov::op::v0::Convert>(downconvert, data_prec);
+
+            result = std::make_shared<ov::op::v1::Divide>(upconvert, input_scale);
+        } else {
+            const auto shift = std::make_shared<ov::op::v1::Subtract>(scale, input_shift);
+
+            const auto downconvert = std::make_shared<ov::op::v0::Convert>(shift, dst_prec);
+            const auto upconvert = std::make_shared<ov::op::v0::Convert>(downconvert, data_prec);
+
+            const auto deshift = std::make_shared<ov::op::v1::Add>(upconvert, input_shift);
+            result = std::make_shared<ov::op::v1::Divide>(deshift, input_scale);
+        }
+
+        model_ref = std::make_shared<ov::Model>(NodeVector{result}, params);
+    }
+
+    const auto res = compare_functions(model, model_ref);
+    ASSERT_TRUE(res.first) << res.second;
+}
+
+const std::vector<element::Type_t> data_precisions = {element::Type_t::f32,
+                                                      element::Type_t::f16,
+                                                      element::Type_t::bf16};
+
+const std::vector<element::Type_t> destination_precisions = {element::Type_t::f8e4m3, element::Type_t::f8e5m2};
+
+const std::vector<bool> default_shift = {true, false};
+
+const auto simple_fake_convert_params = ::testing::Combine(::testing::Values(Shape{2, 3, 4, 5}),
+                                                           ::testing::Values(Shape{1}),
+                                                           ::testing::Values(Shape{1}),
+                                                           ::testing::ValuesIn(data_precisions),
+                                                           ::testing::ValuesIn(destination_precisions),
+                                                           ::testing::ValuesIn(default_shift));
+
+const auto broadcast_fake_convert_params = ::testing::Combine(::testing::Values(Shape{2, 3, 4, 5}),
+                                                              ::testing::Values(Shape{2, 3, 1, 1}),
+                                                              ::testing::Values(Shape{2, 3, 1, 1}),
+                                                              ::testing::ValuesIn(data_precisions),
+                                                              ::testing::ValuesIn(destination_precisions),
+                                                              ::testing::ValuesIn(default_shift));
+
+const auto elementwise_fake_convert_params = ::testing::Combine(::testing::Values(Shape{2, 3, 4, 5}),
+                                                                ::testing::Values(Shape{2, 3, 4, 5}),
+                                                                ::testing::Values(Shape{2, 3, 4, 5}),
+                                                                ::testing::ValuesIn(data_precisions),
+                                                                ::testing::ValuesIn(destination_precisions),
+                                                                ::testing::ValuesIn(default_shift));
+
+INSTANTIATE_TEST_SUITE_P(SimpleFakeConvert_Decomposition,
+                         FakeConvertDecompositionTest,
+                         simple_fake_convert_params,
+                         FakeConvertDecompositionTest::getTestCaseName);
+
+INSTANTIATE_TEST_SUITE_P(BroadcastFakeConvert_Decomposition,
+                         FakeConvertDecompositionTest,
+                         broadcast_fake_convert_params,
+                         FakeConvertDecompositionTest::getTestCaseName);
+
+INSTANTIATE_TEST_SUITE_P(ElementwiseFakeConvert_Decomposition,
+                         FakeConvertDecompositionTest,
+                         elementwise_fake_convert_params,
+                         FakeConvertDecompositionTest::getTestCaseName);

src/plugins/intel_cpu/src/transformations/transformation_pipeline.cpp

@@ -80,6 +80,7 @@
 #include "transformations/op_conversions/detection_output_downgrade.hpp"
 #include "transformations/op_conversions/detection_output_upgrade.hpp"
 #include "transformations/op_conversions/eye_decomposition.hpp"
+#include "transformations/op_conversions/fake_convert_decomposition.hpp"
 #include "transformations/op_conversions/fq_decomposition.hpp"
 #include "transformations/op_conversions/gelu7_downgrade.hpp"
 #include "transformations/op_conversions/group_normalization_decomposition.hpp"
@@ -1293,6 +1294,7 @@ void Transformations::PostSnippets(void) {
             return node::FakeQuantize::isSupportedOperation(node, errMsg);
         },
         ov::pass::FakeQuantizeDecomposition);
+    CPU_REGISTER_PASS_COMMON(postSnippetsManager, ov::pass::FakeConvertDecomposition);
     CPU_REGISTER_PASS_COMMON(postSnippetsManager, ov::pass::ConstantFolding);
     postSnippetsManager.run_passes(model);
 }

src/plugins/intel_cpu/tests/functional/shared_tests_instances/single_layer_tests/fake_convert.cpp

@@ -0,0 +1,59 @@
+// Copyright (C) 2024 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#include "single_op_tests/fake_convert.hpp"
+
+namespace {
+using ov::test::FakeConvertLayerTest;
+
+const std::vector<std::vector<ov::Shape>> shapes = {{{2, 3, 4, 5}}};
+
+const std::vector<ov::element::Type> data_precisions = {ov::element::f32, ov::element::f16, ov::element::bf16};
+
+const std::vector<ov::element::Type> destination_precisions = {ov::element::f8e4m3, ov::element::f8e5m2};
+
+const std::vector<bool> default_shift = {true, false};
+
+const auto simple_fake_convert_params =
+    ::testing::Combine(::testing::ValuesIn(ov::test::static_shapes_to_test_representation(shapes)),
+                       ::testing::Values(ov::Shape{1}),
+                       ::testing::Values(ov::Shape{1}),
+                       ::testing::ValuesIn(data_precisions),
+                       ::testing::ValuesIn(destination_precisions),
+                       ::testing::ValuesIn(default_shift),
+                       ::testing::Values(ov::test::utils::DEVICE_CPU));
+
+const auto broadcast_fake_convert_params =
+    ::testing::Combine(::testing::ValuesIn(ov::test::static_shapes_to_test_representation(shapes)),
+                       ::testing::Values(ov::Shape{2, 3, 1, 1}),
+                       ::testing::Values(ov::Shape{2, 3, 1, 1}),
+                       ::testing::ValuesIn(data_precisions),
+                       ::testing::ValuesIn(destination_precisions),
+                       ::testing::ValuesIn(default_shift),
+                       ::testing::Values(ov::test::utils::DEVICE_CPU));
+
+const auto elementwise_fake_convert_params =
+    ::testing::Combine(::testing::ValuesIn(ov::test::static_shapes_to_test_representation(shapes)),
+                       ::testing::Values(ov::Shape{2, 3, 4, 5}),
+                       ::testing::Values(ov::Shape{2, 3, 4, 5}),
+                       ::testing::ValuesIn(data_precisions),
+                       ::testing::ValuesIn(destination_precisions),
+                       ::testing::ValuesIn(default_shift),
+                       ::testing::Values(ov::test::utils::DEVICE_CPU));
+
+INSTANTIATE_TEST_SUITE_P(smoke_FakeConvert_simple,
+                         FakeConvertLayerTest,
+                         simple_fake_convert_params,
+                         FakeConvertLayerTest::getTestCaseName);
+
+INSTANTIATE_TEST_SUITE_P(smoke_FakeConvert_broadcast,
+                         FakeConvertLayerTest,
+                         broadcast_fake_convert_params,
+                         FakeConvertLayerTest::getTestCaseName);
+
+INSTANTIATE_TEST_SUITE_P(smoke_FakeConvert_elementwise,
+                         FakeConvertLayerTest,
+                         elementwise_fake_convert_params,
+                         FakeConvertLayerTest::getTestCaseName);
+}  // namespace

src/tests/functional/plugin/shared/include/single_op_tests/fake_convert.hpp

@@ -0,0 +1,16 @@
+// Copyright (C) 2024 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#pragma once
+
+#include "shared_test_classes/single_op/fake_convert.hpp"
+
+namespace ov {
+namespace test {
+
+TEST_P(FakeConvertLayerTest, Inference) {
+    run();
+}
+}  // namespace test
+}  // namespace ov

src/tests/functional/shared_test_classes/include/shared_test_classes/single_op/fake_convert.hpp

@@ -0,0 +1,28 @@
+// Copyright (C) 2024 Intel Corporation
+// SPDX-License-Identifier: Apache-2.0
+//
+
+#pragma once
+
+#include "shared_test_classes/base/ov_subgraph.hpp"
+
+namespace ov {
+namespace test {
+using FakeConvertParams = std::tuple<std::vector<InputShape>,  // Data shape
+                                     Shape,                    // Scale shape
+                                     Shape,                    // Shift shape
+                                     ov::element::Type,        // Input precision
+                                     ov::element::Type,        // Ddestination precision
+                                     bool,                     // Default shift
+                                     std::string>;             // Device name
+
+class FakeConvertLayerTest : public testing::WithParamInterface<FakeConvertParams>,
+                             virtual public ov::test::SubgraphBaseTest {
+public:
+    static std::string getTestCaseName(const testing::TestParamInfo<FakeConvertParams>& obj);
+
+protected:
+    void SetUp() override;
+};
+}  // namespace test
+}  // namespace ov