[IE Transformations] Disable pull_transpose_through_fq transformation for activations path (#8178)

* [IE Transformations] Disable pull_transpose_through_fq transformation for activations path

* Update MO part

* Replace TEST to TEST_F

* Fix tests with fixture.

Remove dynamic shapes tests due to the fact that there is always a constant at the input

inference-engine/src/transformations/src/transformations/common_optimizations/pull_transpose_through_fq.cpp

@@ -17,7 +17,8 @@ NGRAPH_RTTI_DEFINITION(ngraph::pass::PullTransposeThroughFQUp, "PullTransposeThr
 
 ngraph::pass::PullTransposeThroughFQUp::PullTransposeThroughFQUp() {
     MATCHER_SCOPE(PullTransposeThroughFQUp);
-    auto m_fq = pattern::wrap_type<opset1::FakeQuantize>({pattern::any_input(pattern::has_static_rank()),
+    const auto weights = ngraph::pattern::wrap_type<ngraph::opset1::Constant>();
+    auto m_fq = pattern::wrap_type<opset1::FakeQuantize>({weights,
                                                           pattern::any_input(pattern::has_static_shape()),
                                                           pattern::any_input(pattern::has_static_shape()),
                                                           pattern::any_input(pattern::has_static_shape()),

inference-engine/tests/functional/inference_engine/transformations/ngraph_fq_transpose_test.cpp

@@ -56,28 +56,28 @@ TEST_F(TransformationTestsF, FQTransposeTest1) {
     }
 }
 
-TEST(TransformationTests, FQTransposeDynamic) {
+TEST_F(TransformationTestsF, FQTransposeNegativeCase) {
-    auto data = std::make_shared<ngraph::op::Parameter>(ngraph::element::f32, ngraph::PartialShape::dynamic());
+    auto create_graph = []() -> std::shared_ptr<ngraph::Function> {
-    auto input_low = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {2});
+        auto data = std::make_shared<ngraph::op::Parameter>(ngraph::element::f32, ngraph::PartialShape{1, 3, 1});
-    auto input_high = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {3});
+        auto sigmoid = std::make_shared<ngraph::op::Sigmoid>(data);
-    auto output_low = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {2});
+        auto input_low = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {2});
-    auto output_high = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {3});
+        auto input_high = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {3});
-    auto transpose_order = ngraph::op::Constant::create(ngraph::element::i64, ngraph::Shape{3}, {0, 2, 1});
+        auto output_low = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {2});
+        auto output_high = ngraph::op::Constant::create(ngraph::element::f32, ngraph::Shape{1}, {3});
+        auto transpose_order = ngraph::op::Constant::create(ngraph::element::i64, ngraph::Shape{3}, {0, 2, 1});
 
-    std::shared_ptr<ngraph::Function> f(nullptr);
+        auto fq = std::make_shared<ngraph::op::FakeQuantize>(sigmoid, input_low, input_high, output_low, output_high, 1);
-    {
-        auto fq = std::make_shared<ngraph::op::FakeQuantize>(data, input_low, input_high, output_low, output_high, 1);
         auto transpose = std::make_shared<ngraph::op::Transpose>(fq, transpose_order);
 
-        f = std::make_shared<ngraph::Function>(ngraph::NodeVector{transpose}, ngraph::ParameterVector{data});
+        return std::make_shared<ngraph::Function>(ngraph::NodeVector{transpose}, ngraph::ParameterVector{data});
+    };
+    function = create_graph();
 
-        ngraph::pass::Manager manager;
+    manager.register_pass<ngraph::pass::InitNodeInfo>();
-        manager.register_pass<ngraph::pass::InitNodeInfo>();
+    manager.register_pass<ngraph::pass::PullTransposeThroughFQUp>();
-        manager.register_pass<ngraph::pass::PullTransposeThroughFQUp>();
+    manager.register_pass<ngraph::pass::InjectionPass>([](std::shared_ptr<ngraph::Function> f) {
-        manager.register_pass<ngraph::pass::InjectionPass>([](std::shared_ptr<ngraph::Function> f) {
+        check_rt_info(f);
-            check_rt_info(f);
+    });
-        });
+
-        manager.register_pass<ngraph::pass::ConstantFolding>();
+    function_ref = create_graph();
-        ASSERT_NO_THROW(manager.run_passes(f));
-    }
 }

model-optimizer/extensions/back/MatMulNormalizer.py

@@ -73,14 +73,13 @@ class MatMulConstTransposesExtraction(BackReplacementPattern):
 class PullTransposeThroughFQUp(BackReplacementPattern):
     r"""
         BEFORE                                      AFTER
-                                                        T  T T  T  T
+    Const                                             Const
-         \ \ | / /                                       \ \ | / /
+        \ \ | / /                                       |
-        FakeQuantize                                    FakeQuantize
+        FakeQuantize                                    T  T T  T  T
+            |                                            \ \ | / /
+        Transpose                                       FakeQuantize
             |                                                |
-        Transpose                                         next_op
+         next_op                                          next_op
-            |
-         next_op
-
         `T` is Transpose for short
     """
     enabled = True
@@ -94,13 +93,17 @@ class PullTransposeThroughFQUp(BackReplacementPattern):
     def pattern():
         return dict(
             nodes=[
+                ('fq_const_input', dict(kind='op', type='Const')),
+                ('fq_const_input_d', dict()),
                 ('fq', dict(kind='op', type='FakeQuantize')),
-                ('data', dict()),
+                ('fq_d', dict()),
                 ('transpose', dict(kind='op', type='Transpose')),
             ],
             edges=[
-                ('fq', 'data'),
+                ('fq_const_input', 'fq_const_input_d'),
-                ('data', 'transpose'),
+                ('fq_const_input_d', 'fq', {'in': 0}),
+                ('fq', 'fq_d'),
+                ('fq_d', 'transpose'),
             ]
         )
 

model-optimizer/unit_tests/extensions/back/MatMulNormalizer_test.py

@@ -15,7 +15,7 @@ from mo.front.common.partial_infer.utils import int64_array
 from mo.ops.reshape import Reshape
 from mo.utils.ir_engine.compare_graphs import compare_graphs
 from unit_tests.utils.graph import build_graph, regular_op_with_shaped_data, valued_const_with_data, \
-    result, connect, connect_data
+    shaped_const_with_data, result, connect, connect_data
 from unit_tests.utils.graph import regular_op_with_empty_data as op_with_empty_data
 
 
@@ -101,9 +101,8 @@ class SmartReshape_HC_Reshape_MatMulTest(unittest.TestCase):
 
 
 class FQTransposePullerTest(unittest.TestCase):
-    def nodes(self, input_shape, transpose_shape, fq_shape):
+    def nodes(self, input_shape, transpose_shape, fq_shape, is_input_const):
-        return {
+        nodes = {
-            **regular_op_with_shaped_data('input', input_shape, dict(type='Parameter', op='Parameter')),
             **valued_const_with_data('il', np.array([[[[0]]]])),
             **valued_const_with_data('ih', np.array([[[[255]]]])),
             **valued_const_with_data('ol', np.array([[[[0]]]])),
@@ -116,8 +115,16 @@ class FQTransposePullerTest(unittest.TestCase):
             **result(),
         }
 
+        if is_input_const:
+            input_node = shaped_const_with_data('input', input_shape)
+        else:
+            input_node = regular_op_with_shaped_data('input', input_shape, dict(type='Parameter', op='Parameter'))
+
+        nodes.update(input_node)
+        return nodes
+
     def test_positive(self):
-        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 3, 224, 224])
+        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 3, 224, 224], True)
         edges = [
             *connect('input', '0:FQ'),
             *connect('il', '1:FQ'),
@@ -132,7 +139,7 @@ class FQTransposePullerTest(unittest.TestCase):
         PullTransposeThroughFQUp().find_and_replace_pattern(graph)
         graph.clean_up()
 
-        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 224, 224, 3])
+        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 224, 224, 3], True)
         edges = [
             *connect('input', '0:transpose'),
             *connect('order:0', '1:transpose'),
@@ -148,8 +155,8 @@ class FQTransposePullerTest(unittest.TestCase):
         (flag, resp) = compare_graphs(graph, graph_ref, 'output', check_op_attrs=True)
         self.assertTrue(flag, resp)
 
-    def test_negative(self):
+    def test_negative_1(self):
-        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 3, 224, 224])
+        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 3, 224, 224], True)
         edges = [
             *connect('input', '0:FQ'),
             *connect('il', '1:FQ'),
@@ -168,3 +175,21 @@ class FQTransposePullerTest(unittest.TestCase):
         (flag, resp) = compare_graphs(graph, graph_ref, 'output', check_op_attrs=True)
         self.assertTrue(flag, resp)
 
+    def test_negative_2(self):
+        nodes = self.nodes([1, 3, 224, 224], [1, 224, 224, 3], [1, 3, 224, 224], False)
+        edges = [
+            *connect('input', '0:FQ'),
+            *connect('il', '1:FQ'),
+            *connect('ih', '2:FQ'),
+            *connect('ol', '3:FQ'),
+            *connect('oh', '4:FQ'),
+            *connect('FQ:0', '0:transpose'),
+            *connect('order:0', '1:transpose'),
+            *connect('transpose:0', 'output'),
+        ]
+        graph = build_graph(nodes_attrs=nodes, edges=edges, nodes_with_edges_only=True)
+        graph_ref = graph.copy()
+        PullTransposeThroughFQUp().find_and_replace_pattern(graph)
+
+        (flag, resp) = compare_graphs(graph, graph_ref, 'output', check_op_attrs=True)
+        self.assertTrue(flag, resp)