IVGCVSW-6252 Armnn Error: Failed to parse operator #1 within subgraph #0 error: Operator not supported

* Added missing support for reshape operator in tflite parser when the target shape is dynamic and batch size is unknown
* Added corresponding unit test

Change-Id: I78f4b386f2e5e0d08ecacde17e33e304f6afbb70
Signed-off-by: Tamas Nyiri <tamas.nyiri@arm.com>

3 files changed, 215 insertions, 9 deletions

diff --git a/CMakeLists.txt b/CMakeLists.txt
index fde058216b..d08641431d 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -718,6 +718,7 @@ if(BUILD_UNIT_TESTS)
              src/armnnTfLiteParser/test/Prelu.cpp
              src/armnnTfLiteParser/test/Reduce.cpp
              src/armnnTfLiteParser/test/Reshape.cpp
+             src/armnnTfLiteParser/test/ReshapeDynamic.cpp
              src/armnnTfLiteParser/test/ResizeBilinear.cpp
              src/armnnTfLiteParser/test/ResizeNearestNeighbor.cpp
              src/armnnTfLiteParser/test/Quantize.cpp
diff --git a/src/armnnTfLiteParser/TfLiteParser.cpp b/src/armnnTfLiteParser/TfLiteParser.cpp
index 15ca36d906..4e0efbbf3c 100644
--- a/src/armnnTfLiteParser/TfLiteParser.cpp
+++ b/src/armnnTfLiteParser/TfLiteParser.cpp
@@ -2545,9 +2545,11 @@ void TfLiteParserImpl::ParseReshape(size_t subgraphIndex, size_t operatorIndex)
     // If there is no built-in option given or if the built-in new_shape parameter was empty
     if (!targetShapeFound)
     {
-        // Check for a second input tensor
-        if (inputs.size() > 1 && inputs[1] != nullptr)
+        bool secondInputTensorExists = inputs.size() > 1 && inputs[1] != nullptr;
+        if (secondInputTensorExists)
         {
+            armnn::TensorInfo shapeInfo  = ToTensorInfo(inputs[1]);
+
             if (inputs[1]->is_variable)
             {
                 ARMNN_THROW_PARSE_EXCEPTION( "Target shapes defined in non-const input tensors is not supported");
@@ -2563,16 +2565,47 @@ void TfLiteParserImpl::ParseReshape(size_t subgraphIndex, size_t operatorIndex)
                 ARMNN_THROW_PARSE_EXCEPTION("Target 'shape' input is not an int32 type");
             }
 
-            // Extract target shape from input
-            auto bufferPtr = GetBuffer(m_Model, inputs[1]->buffer);
-            auto values = reinterpret_cast<const int32_t*>(bufferPtr->data.data());
-            if (!values)
+            // If the provided shape is constant, we can load in from the buffer ...
+            if (shapeInfo.IsConstant())
             {
-                ARMNN_THROW_PARSE_EXCEPTION("Reshape operator target shape input buffer data is null");
+                // Extract target shape from input
+                auto bufferPtr = GetBuffer(m_Model, inputs[1]->buffer);
+                auto values = reinterpret_cast<const int32_t*>(bufferPtr->data.data());
+                if (!values)
+                {
+                    ARMNN_THROW_PARSE_EXCEPTION("Reshape operator target shape input buffer data is null");
+                }
+                for (int i=0; i < inputs[1]->shape[0]; ++i)
+                {
+                    targetShape.push_back(values[i]);
+                }
             }
-            for (int i=0; i < inputs[1]->shape[0]; ++i)
+            // ... otherwise, we have to infer it during Runtime.
+            else
             {
-                targetShape.push_back(values[i]);
+                // The parser only supports shape (batch, -1) or (-1) for non-constant shape input.
+                unsigned int dims = shapeInfo.GetNumDimensions();
+                TensorShape shapes = shapeInfo.GetShape();
+                if (dims != 1 || shapes[0] > 2)
+                {
+                    throw ParseException(fmt::format("Invalid input shape '{}' in Reshape layer '{}' {}. "
+                                                    "The parser only supports shape (batch, -1) or (-1) for "
+                                                    "non-constant shape input.",
+                                                    shapes[0],
+                                                    layerName,
+                                                    CHECK_LOCATION().AsString()));
+                }
+
+                const int32_t numInputElements = inputTensorInfo.GetNumElements();
+                const int32_t inputTensorShape = inputTensorInfo.GetShape()[0];
+                if (shapes[0] == 1)
+                {
+                    targetShape = { numInputElements };
+                }
+                else if (shapes[0] == 2)
+                {
+                    targetShape = { inputTensorShape , numInputElements / inputTensorShape };
+                }
             }
         }
         else
diff --git a/src/armnnTfLiteParser/test/ReshapeDynamic.cpp b/src/armnnTfLiteParser/test/ReshapeDynamic.cpp
new file mode 100644
index 0000000000..ca993fb90c
--- /dev/null
+++ b/src/armnnTfLiteParser/test/ReshapeDynamic.cpp
@@ -0,0 +1,172 @@
+//
+// Copyright © 2021 Arm Ltd and Contributors. All rights reserved.
+// SPDX-License-Identifier: MIT
+//
+
+#include "ParserFlatbuffersFixture.hpp"
+
+
+TEST_SUITE("TensorflowLiteParser_Reshape_Dynamic")
+{
+struct ReshapeDynamicFixture : public ParserFlatbuffersFixture
+{
+    explicit ReshapeDynamicFixture()
+    {
+        m_JsonString = R"(
+{
+  "version": 3,
+  "operator_codes": [
+    {
+      "deprecated_builtin_code": 77,
+      "version": 1,
+      "builtin_code": "ADD"
+    },
+    {
+      "deprecated_builtin_code": 22,
+      "version": 1,
+      "builtin_code": "ADD"
+    }
+  ],
+  "subgraphs": [
+    {
+      "tensors": [
+        {
+          "shape": [
+            1,
+            9
+          ],
+          "type": "FLOAT32",
+          "buffer": 1,
+          "name": "input_33",
+          "quantization": {
+            "details_type": "NONE",
+            "quantized_dimension": 0
+          },
+          "is_variable": false,
+          "shape_signature": [
+            -1,
+            9
+          ]
+        },
+        {
+          "shape": [
+            2
+          ],
+          "type": "INT32",
+          "buffer": 2,
+          "name": "functional_15/tf_op_layer_Shape_9/Shape_9",
+          "quantization": {
+            "details_type": "NONE",
+            "quantized_dimension": 0
+          },
+          "is_variable": false
+        },
+        {
+          "shape": [
+            1,
+            9
+          ],
+          "type": "FLOAT32",
+          "buffer": 3,
+          "name": "Identity",
+          "quantization": {
+            "details_type": "NONE",
+            "quantized_dimension": 0
+          },
+          "is_variable": false,
+          "shape_signature": [
+            -1,
+            9
+          ]
+        }
+      ],
+      "inputs": [
+        0
+      ],
+      "outputs": [
+        2
+      ],
+      "operators": [
+        {
+          "opcode_index": 0,
+          "inputs": [
+            0
+          ],
+          "outputs": [
+            1
+          ],
+          "builtin_options_type": "ShapeOptions",
+          "builtin_options": {
+            "out_type": "INT32"
+          },
+          "custom_options_format": "FLEXBUFFERS"
+        },
+        {
+          "opcode_index": 1,
+          "inputs": [
+            0,
+            1
+          ],
+          "outputs": [
+            2
+          ],
+          "builtin_options_type": "NONE",
+          "custom_options_format": "FLEXBUFFERS"
+        }
+      ],
+      "name": "main"
+    }
+  ],
+  "description": "MLIR Converted.",
+  "buffers": [
+    {
+    },
+    {
+    },
+    {
+    },
+    {
+    },
+    {
+      "data": [
+        49,
+        46,
+        49,
+        48,
+        46,
+        48,
+        0,
+        0,
+        0,
+        0,
+        0,
+        0,
+        0,
+        0,
+        0,
+        0
+      ]
+    }
+  ],
+  "metadata": [
+    {
+      "name": "min_runtime_version",
+      "buffer": 4
+    }
+  ]
+}
+)";
+
+    }
+};
+
+TEST_CASE_FIXTURE(ReshapeDynamicFixture, "ParseReshapeDynamic")
+{
+    SetupSingleInputSingleOutput("input_33", "Identity");
+    RunTest<2, armnn::DataType::Float32>(0,
+                                                 { 1, 2, 3, 4, 5, 6, 7, 8, 9 },
+                                                 { 1, 2, 3, 4, 5, 6, 7, 8, 9 });
+    CHECK((m_Parser->GetNetworkOutputBindingInfo(0, "Identity").second.GetShape()
+                == armnn::TensorShape({1,9})));
+}
+}
\ No newline at end of file