Implement reshape in VM for tensors with rank<=2 (#184)

* Type checking for reshape op

* Formatting

* Implement reduce op for matrices and smaller in VM

* Formatting

* Missing include for Ubuntu

* Remove IndexBuiltin class

src/ngraph/CMakeLists.txt

@@ -36,6 +36,7 @@ set (SRC
     ops/op.cpp
     ops/parameter.cpp
     ops/reduce.cpp
+    ops/reshape.cpp
     ops/select.cpp
     ops/tuple.cpp
     ops/unary_elementwise_arithmetic.cpp

src/ngraph/ngraph.hpp

@@ -77,6 +77,7 @@
 #include "ngraph/ops/power.hpp"
 #include "ngraph/ops/reduce.hpp"
 #include "ngraph/ops/remainder.hpp"
+#include "ngraph/ops/reshape.hpp"
 #include "ngraph/ops/select.hpp"
 #include "ngraph/ops/subtract.hpp"
 #include "ngraph/ops/tuple.hpp"

src/ngraph/ops/broadcast.hpp

@@ -20,7 +20,7 @@ namespace ngraph
 {
     namespace op
     {
-        class Broadcast : public IndexBuiltin
+        class Broadcast : public Builtin
         {
         public:
             ///
@@ -32,7 +32,7 @@ namespace ngraph
             Broadcast(const std::shared_ptr<Node>& arg,
                       const Shape& shape,
                       const AxisSet& broadcast_axes)
-                : IndexBuiltin(arg)
+                : Builtin({arg})
                 , m_shape(shape)
                 , m_broadcast_axes(broadcast_axes)
             {

src/ngraph/ops/op.hpp

@@ -38,31 +38,6 @@ namespace ngraph
             }
         };
 
-        /// Index ops create a new way to index the same tensor elements
-        class IndexBuiltin : public Builtin
-        {
-        protected:
-            IndexBuiltin(const std::shared_ptr<Node>& arg)
-                : Builtin(Nodes{arg})
-            {
-            }
-        };
-
-        class Reshape : public IndexBuiltin
-        {
-        public:
-            Reshape(const std::shared_ptr<Node>& arg0, const Shape& shape)
-                : IndexBuiltin(arg0)
-                , m_shape(shape)
-            {
-            }
-
-            virtual std::string description() const override { return "Reshape"; }
-            //virtual void propagate_types() override;
-        protected:
-            Shape m_shape;
-        };
-
         /// Operations where the same element function is applied to each element
         /// Op(X)[I] = op(X[I])
         class UnaryElementwiseBuiltin : public Builtin

src/ngraph/ops/reshape.cpp

+// ----------------------------------------------------------------------------
+// Copyright 2017 Nervana Systems Inc.
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// ----------------------------------------------------------------------------
+
+#include "ngraph/ops/reshape.hpp"
+#include "ngraph/function.hpp"
+
+#include <algorithm>
+
+using namespace std;
+using namespace ngraph::op;
+
+void Reshape::propagate_types()
+{
+    if (m_arguments.size() != 1)
+    {
+        throw ngraph_error("Wrong number of arguments.");
+    }
+
+    auto arg_type = m_arguments.at(0)->get_value_type();
+    if (nullptr == arg_type)
+    {
+        throw ngraph_error("Argument to reshape is missing type.");
+    }
+    auto arg_tensor_view_type = dynamic_pointer_cast<const TensorViewType>(arg_type);
+    if (nullptr == arg_type)
+    {
+        throw ngraph_error("Argument to reshape is not a tensor view");
+    }
+
+    auto arg_shape = arg_tensor_view_type->get_shape();
+    auto arg_rank = arg_shape.size();
+
+    if (m_input_order.size() != arg_rank)
+    {
+        throw ngraph_error("Input axis order for reshape is not a permutation of argument's axes");
+    }
+
+    for (size_t i = 0; i < arg_rank; i++)
+    {
+        auto it = std::find(std::begin(m_input_order), std::end(m_input_order), i);
+        if (std::end(m_input_order) == it)
+        {
+            throw ngraph_error(
+                "Input axis order for reshape is not a permutation of argument's axes");
+        }
+    }
+
+    size_t arg_shape_product = 1;
+    for (auto i : arg_shape)
+    {
+        arg_shape_product *= i;
+    }
+
+    size_t output_shape_product = 1;
+    for (auto i : m_output_shape)
+    {
+        output_shape_product *= i;
+    }
+
+    if (arg_shape_product != output_shape_product)
+    {
+        throw ngraph_error(
+            "Product of output shape dimensions does not match product of argument shape "
+            "dimensions for reshape");
+    }
+
+    set_value_type_checked(
+        make_shared<TensorViewType>(arg_tensor_view_type->get_element_type(), m_output_shape));
+}

src/ngraph/ops/reshape.hpp

+// ----------------------------------------------------------------------------
+// Copyright 2017 Nervana Systems Inc.
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// ----------------------------------------------------------------------------
+
+#pragma once
+
+#include "ngraph/ops/op.hpp"
+
+namespace ngraph
+{
+    namespace op
+    {
+        class Reshape : public Builtin
+        {
+        public:
+            ///
+            /// @param arg The tensor view to be reshaped.
+            /// @param input_order The order in which to iterate over input axes. (TODO: that needs more explanation)
+            ///        This must be a permutation of the sequence (0,...,n-1) where n is the rank of the input tensor.
+            /// @param output_shape The output shape. If the input shape is (a0,...,ak-1) then the output shape must
+            ///        be of the form (b0,...,bj-1) where product(ai) == product(bi).
+            ///
+            Reshape(const std::shared_ptr<Node>& arg,
+                    const AxisVector& input_order,
+                    const Shape& output_shape)
+                : Builtin({arg})
+                , m_input_order(input_order)
+                , m_output_shape(output_shape)
+            {
+            }
+
+            virtual std::string description() const override { return "Reshape"; }
+            virtual void propagate_types() override;
+
+            const AxisVector& get_input_order() const { return m_input_order; }
+            const Shape& get_output_shape() const { return m_output_shape; }
+        protected:
+            const AxisVector m_input_order;
+            const Shape m_output_shape;
+        };
+    }
+}

src/ngraph/runtime/ngvm/eigen/matrix_transpose.hpp

+// ----------------------------------------------------------------------------
+// Copyright 2017 Nervana Systems Inc.
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//      http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// ----------------------------------------------------------------------------
+
+#pragma once
+
+#include "ngraph/runtime/ngvm/call_frame.hpp"
+#include "ngraph/runtime/ngvm/eigen/utils.hpp"
+#include "ngraph/runtime/ngvm/instruction.hpp"
+#include "ngraph/runtime/tensor_view.hpp"
+
+namespace ngraph
+{
+    namespace runtime
+    {
+        namespace ngvm
+        {
+            namespace eigen
+            {
+                template <typename ET>
+                class MatrixTransposeInstruction : public Instruction
+                {
+                public:
+                    MatrixTransposeInstruction(const TensorViewInfo& arg, const TensorViewInfo& out)
+                        : m_arg(arg)
+                        , m_out(out)
+                    {
+                    }
+
+                    virtual void execute(CallFrame& call_frame) const override
+                    {
+                        EigenMatrix<ET>(call_frame, m_out) =
+                            EigenMatrix<ET>(call_frame, m_arg).transpose();
+                    }
+
+                protected:
+                    TensorViewInfo m_arg;
+                    TensorViewInfo m_out;
+                };
+            }
+        }
+    }
+}

src/ngraph/runtime/ngvm/external_function.cpp

@@ -44,6 +44,7 @@
 #include "ngraph/ops/negative.hpp"
 #include "ngraph/ops/not_equal.hpp"
 #include "ngraph/ops/reduce.hpp"
+#include "ngraph/ops/reshape.hpp"
 #include "ngraph/ops/select.hpp"
 #include "ngraph/ops/subtract.hpp"
 #include "ngraph/ops/tuple.hpp"
@@ -71,6 +72,7 @@
 #include "ngraph/runtime/ngvm/eigen/less_than.hpp"
 #include "ngraph/runtime/ngvm/eigen/log.hpp"
 #include "ngraph/runtime/ngvm/eigen/matrix_mult.hpp"
+#include "ngraph/runtime/ngvm/eigen/matrix_transpose.hpp"
 #include "ngraph/runtime/ngvm/eigen/matrix_vector_product.hpp"
 #include "ngraph/runtime/ngvm/eigen/maximum.hpp"
 #include "ngraph/runtime/ngvm/eigen/multiply.hpp"
@@ -777,6 +779,59 @@ ExternalFunction::OpMap& ExternalFunction::get_op_map()
             }
         };
 
+        REGISTER_TO_OP_MAP(op::Reshape)
+        {
+            auto reshape = static_cast<const op::Reshape*>(n);
+
+            auto arg_type = reshape->get_arguments().at(0)->get_value_type();
+            auto arg_tensor_view_type = dynamic_pointer_cast<const TensorViewType>(arg_type);
+            assert(nullptr != arg_tensor_view_type);
+            auto arg_shape = arg_tensor_view_type->get_shape();
+            auto arg_rank = arg_shape.size();
+
+            auto result_type = reshape->get_value_type();
+            auto result_tensor_view_type = dynamic_pointer_cast<const TensorViewType>(result_type);
+            assert(nullptr != result_tensor_view_type);
+            auto result_shape = result_tensor_view_type->get_shape();
+            auto& result_element_type = result_tensor_view_type->get_element_type();
+
+            auto input_order = reshape->get_input_order();
+
+            bool same_layout = std::is_sorted(input_order.begin(), input_order.end());
+
+            size_t result_shape_product = 1;
+            for (auto i : result_shape)
+            {
+                result_shape_product *= i;
+            }
+
+            // If there is no layout change or we are just going from 1^n to 1^m or a zero-size tensor, we can just copy.
+            if (same_layout || result_shape_product < 2)
+            {
+                PUSH_POLYMORPHIC_INSTRUCTION(result_element_type,
+                                             "Reshape has unhandled element type",
+                                             runtime::ngvm::eigen::CopyInstruction,
+                                             in.at(0).get_index(),
+                                             out.at(0).get_index());
+            }
+            // If there *is* a layout change in the 2D case, we transpose the input.
+            else if (arg_rank == 2)
+            {
+                PUSH_POLYMORPHIC_INSTRUCTION(result_element_type,
+                                             "Reshape has unhandled element type",
+                                             runtime::ngvm::eigen::MatrixTransposeInstruction,
+                                             in[0],
+                                             out[0]);
+            }
+            // Other cases (reordering of axes for tensors with rank>2) are not handled yet.
+            else
+            {
+                throw ngraph_error(
+                    "Axis permutation in reshape is not implemented yet for tensors with rank>2 in "
+                    "VM");
+            }
+        };
+
         initialized = true;
     }
     return op_map;

test/execute.cpp

@@ -1620,3 +1620,243 @@ TEST(execute, reduce_matrix_to_scalar_zero_by_zero)
     ASSERT_EQ((vector<float>{}), a->get_vector());
     ASSERT_EQ((vector<float>{99}), b->get_vector());
 }
+
+TEST(type_prop, reshape_t2v_012)
+{
+    auto shape_a = Shape{2, 2, 3};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{12};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{0, 1, 2}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12}), result->get_vector());
+}
+
+TEST(type_prop, reshape_t2s_012)
+{
+    auto shape_a = Shape{1, 1, 1};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{0, 1, 2}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{6};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{6}), result->get_vector());
+}
+
+TEST(type_prop, reshape_t2s_120)
+{
+    auto shape_a = Shape{1, 1, 1};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{1, 2, 0}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{6};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{6}), result->get_vector());
+}
+
+TEST(type_prop, reshape_s2t)
+{
+    auto shape_a = Shape{};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{1, 1, 1, 1, 1, 1};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{42};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{42}), result->get_vector());
+}
+
+TEST(type_prop, reshape_v2m_col)
+{
+    auto shape_a = Shape{3};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{3, 1};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{0}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 2, 3}), result->get_vector());
+}
+
+TEST(type_prop, reshape_v2m_row)
+{
+    auto shape_a = Shape{3};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{1, 3};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{0}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 2, 3}), result->get_vector());
+}
+
+TEST(type_prop, reshape_v2t_middle)
+{
+    auto shape_a = Shape{3};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{1, 3, 1};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{0}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 2, 3}), result->get_vector());
+}
+
+TEST(type_prop, reshape_m2m_same)
+{
+    auto shape_a = Shape{3, 3};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{3, 3};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{0, 1}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3, 4, 5, 6, 7, 8, 9};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 2, 3, 4, 5, 6, 7, 8, 9}), result->get_vector());
+}
+
+TEST(type_prop, reshape_m2m_transpose)
+{
+    auto shape_a = Shape{3, 3};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{3, 3};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{1, 0}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3, 4, 5, 6, 7, 8, 9};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 4, 7, 2, 5, 8, 3, 6, 9}), result->get_vector());
+}
+
+TEST(type_prop, reshape_m2m_dim_change_transpose)
+{
+    auto shape_a = Shape{3, 2};
+    auto A = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), shape_a));
+    auto shape_r = Shape{2, 3};
+    auto rt = make_shared<TensorViewType>(element::Float32::element_type(), shape_r);
+    auto r = make_shared<op::Reshape>(A, AxisVector{1, 0}, shape_r);
+    auto f = make_shared<Function>(r, rt, op::Parameters{A});
+
+    auto manager = runtime::Manager::get("NGVM");
+    auto external = manager->compile(f);
+    auto backend = manager->allocate_backend();
+    auto cf = backend->make_call_frame(external);
+
+    // Create some tensors for input/output
+    auto a = ngraph::runtime::make_tensor<element::Float32>(shape_a);
+    *a = vector<float>{1, 2, 3, 4, 5, 6};
+    auto result = ngraph::runtime::make_tensor<element::Float32>(shape_r);
+
+    (*cf)({a}, {result});
+    ASSERT_EQ((vector<float>{1, 3, 5, 2, 4, 6}), result->get_vector());
+}

test/type_prop.cpp

@@ -1041,3 +1041,197 @@ TEST(type_prop, function_call_deduce)
     auto r_p_r_vt = r_p_r->get_value_type();
     ASSERT_EQ(*r_p_r_vt, TensorViewType(element::Float32::element_type(), shape));
 }
+
+TEST(type_prop, reshape_deduce_s2v)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{}, Shape{1});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{1}));
+}
+
+TEST(type_prop, reshape_deduce_s2m)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{}, Shape{1, 1});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()),
+              TensorViewType(element::Float32::element_type(), Shape{1, 1}));
+}
+
+TEST(type_prop, reshape_deduce_s2t)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{}, Shape{1, 1, 1});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()),
+              TensorViewType(element::Float32::element_type(), Shape{1, 1, 1}));
+}
+
+TEST(type_prop, reshape_deduce_v2s)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{1}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{0}, Shape{});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{}));
+}
+
+TEST(type_prop, reshape_deduce_m2s)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{1, 1}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{0, 1}, Shape{});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{}));
+}
+
+TEST(type_prop, reshape_deduce_t2s)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{1, 1, 1}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{0, 1, 2}, Shape{});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{}));
+}
+
+TEST(type_prop, reshape_deduce_m2v_01)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{0, 1}, Shape{12});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{12}));
+}
+
+TEST(type_prop, reshape_deduce_m2v_10)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 0}, Shape{12});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{12}));
+}
+
+TEST(type_prop, reshape_deduce_t2v_012)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{0, 1, 2}, Shape{60});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{60}));
+}
+
+TEST(type_prop, reshape_deduce_t2v_120)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 2, 0}, Shape{60});
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{60}));
+}
+
+TEST(type_prop, reshape_deduce_correct_t2v_120)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 2, 0}, Shape{60});
+    r->set_value_type(make_shared<TensorViewType>(element::Float32::element_type(), Shape{60}));
+    r->propagate_types();
+    ASSERT_EQ(*(r->get_value_type()), TensorViewType(element::Float32::element_type(), Shape{60}));
+}
+
+TEST(type_prop, reshape_deduce_not_enough_axes)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 0}, Shape{60});
+    try
+    {
+        r->propagate_types();
+        // Should have thrown, so fail if it didn't
+        FAIL() << "Not enough axes not detected";
+    }
+    catch (const ngraph_error& error)
+    {
+        EXPECT_EQ(
+            error.what(),
+            std::string("Input axis order for reshape is not a permutation of argument's axes"));
+    }
+    catch (...)
+    {
+        FAIL() << "Deduced type check failed for unexpected reason";
+    }
+}
+
+TEST(type_prop, reshape_deduce_too_many_axes)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 2, 0, 3}, Shape{60});
+    try
+    {
+        r->propagate_types();
+        // Should have thrown, so fail if it didn't
+        FAIL() << "Too many axes not detected";
+    }
+    catch (const ngraph_error& error)
+    {
+        EXPECT_EQ(
+            error.what(),
+            std::string("Input axis order for reshape is not a permutation of argument's axes"));
+    }
+    catch (...)
+    {
+        FAIL() << "Deduced type check failed for unexpected reason";
+    }
+}
+
+TEST(type_prop, reshape_deduce_duplicate_axes)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 1, 0}, Shape{60});
+    try
+    {
+        r->propagate_types();
+        // Should have thrown, so fail if it didn't
+        FAIL() << "Too many axes not detected";
+    }
+    catch (const ngraph_error& error)
+    {
+        EXPECT_EQ(
+            error.what(),
+            std::string("Input axis order for reshape is not a permutation of argument's axes"));
+    }
+    catch (...)
+    {
+        FAIL() << "Deduced type check failed for unexpected reason";
+    }
+}
+
+TEST(type_prop, reshape_deduce_wrong_output_shape)
+{
+    auto param = make_shared<op::Parameter>(
+        make_shared<TensorViewType>(element::Float32::element_type(), Shape{3, 4, 5}));
+    auto r = make_shared<op::Reshape>(param, AxisVector{1, 2, 0}, Shape{3, 3, 3});
+    try
+    {
+        r->propagate_types();
+        // Should have thrown, so fail if it didn't
+        FAIL() << "Too many axes not detected";
+    }
+    catch (const ngraph_error& error)
+    {
+        EXPECT_EQ(error.what(),
+                  std::string("Product of output shape dimensions does not match "
+                              "product of argument shape dimensions for reshape"));
+    }
+    catch (...)
+    {
+        FAIL() << "Deduced type check failed for unexpected reason";
+    }
+}