backends/reference: Add ReduceSum operation support

This patch addes ReduceSum operation support for reference backend,
which computes the sum of elements across dimensions of a tensor.

Changelog v1:
- Fix file header descriptions.

Chagelog v2:
- Fix line limit issue.
- Fix type conversion issue.

Changelog v3:
- Remove tabs.
- Modify newly added file headers.

Changelog v4:
- Symbol on header isn't allowed so drop it from newly added file headers.

Changelog v5:
- Remove tabs, fix the use of brackets and align lines correctly.

Changelog v6:
- Add serializer and deserializer support.

Changelog v7:
- Fix build error add missed code.

Changelog v8:
- Rename ReduceSumDecriptor to ReduceDescriptor
    - Update m_KeepDims field data type to bool on ReduceDescriptor
    - Add ReduceOperation field to ReduceDescriptor

- Rename ReduceSumLayer to ReduceLayer
    - Update ReduceLayer to use ReduceDescriptor
    - Update ReduceLayer::ValidateTensorShapesFromInputs() function

- Rename RefReduceSumWokload to RefReduceWorkload
    - Update workload to use ReduceDescriptor
    - Update workload to use Decoders and Encoders

- Remove ReduceSum.hpp and ReduceSum.cpp
- Added Reduce.hpp and Reduce.cpp
     - Move Mean.cpp (which is implementing REDUCE_MEAN) functionality to Reduce.cpp
     - Update RefMeanWorkload to call Reduce function with ReduceOperation::Mean argument

- Remove Mean.hpp and Mean.cpp
- Update the Serializer/Deserializer ArmnnSchema.fbs for ReduceLayer, ReduceDescriptor, and ReduceOperation
- Update Serializer and Deserializer for serializing/parsing ReduceLayer
- Added TfLiter parser Sum test for REDUCE_SUM operator
- Make corresponding changes on front-end and Ref backend to support REDUCE_SUM operator

Changelog v9:
- Fixed build errors.

Change-Id: I8c8e034f3df73f9565b3c18eff51ecca6c542195
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>

1 files changed, 54 insertions, 0 deletions

diff --git a/src/armnnTfLiteParser/TfLiteParser.cpp b/src/armnnTfLiteParser/TfLiteParser.cpp
index 1a1e854395..db60224999 100644
--- a/src/armnnTfLiteParser/TfLiteParser.cpp
+++ b/src/armnnTfLiteParser/TfLiteParser.cpp
@@ -10,6 +10,7 @@
 #include <armnn/Exceptions.hpp>
 #include <armnn/Logging.hpp>
 #include <armnn/Tensor.hpp>
+#include <armnnUtils/TensorUtils.hpp>
 #include <armnn/TypesUtils.hpp>
 #include <armnn/utility/Assert.hpp>
 #include <armnn/utility/IgnoreUnused.hpp>
@@ -580,6 +581,7 @@ TfLiteParser::TfLiteParser(const Optional<ITfLiteParser::TfLiteParserOptions>& o
     m_ParserFunctions[tflite::BuiltinOperator_SQUEEZE]                 = &TfLiteParser::ParseSqueeze;
     m_ParserFunctions[tflite::BuiltinOperator_STRIDED_SLICE]           = &TfLiteParser::ParseStridedSlice;
     m_ParserFunctions[tflite::BuiltinOperator_SUB]                     = &TfLiteParser::ParseSub;
+    m_ParserFunctions[tflite::BuiltinOperator_SUM]                     = &TfLiteParser::ParseSum;
     m_ParserFunctions[tflite::BuiltinOperator_TANH]                    = &TfLiteParser::ParseTanH;
     m_ParserFunctions[tflite::BuiltinOperator_TRANSPOSE]               = &TfLiteParser::ParseTranspose;
     m_ParserFunctions[tflite::BuiltinOperator_TRANSPOSE_CONV]          = &TfLiteParser::ParseTransposeConv;
@@ -2994,6 +2996,58 @@ void TfLiteParser::ParseDepthToSpace(size_t subgraphIndex, size_t operatorIndex)
     RegisterOutputSlots(subgraphIndex, operatorIndex, layer, {outputTensorIndexes[0]});
 }
 
+void TfLiteParser::ParseSum(size_t subgraphIndex, size_t operatorIndex)
+{
+    CHECK_MODEL(m_Model, subgraphIndex, operatorIndex);
+
+    const auto &operatorPtr = m_Model->subgraphs[subgraphIndex]->operators[operatorIndex];
+    const auto *options = operatorPtr->builtin_options.AsReducerOptions();
+
+    auto inputs = GetInputs(m_Model, subgraphIndex, operatorIndex);
+    CHECK_VALID_SIZE(inputs.size(), 2);
+
+    auto outputs = GetOutputs(m_Model, subgraphIndex, operatorIndex);
+    CHECK_VALID_SIZE(outputs.size(), 1);
+
+    auto layerName = fmt::format("Sum:{}:{}", subgraphIndex, operatorIndex);
+
+    armnn::TensorInfo inputTensorInfo0 = ToTensorInfo(inputs[0]);
+    armnn::TensorInfo inputTensorInfo1 = ToTensorInfo(inputs[1]);
+    TensorShape input0Shape = inputTensorInfo0.GetShape();
+
+    ReduceDescriptor desc;
+
+    BufferRawPtr axisBufferPtr = GetBuffer(m_Model, inputs[1]->buffer);
+    // Get const axis value from model and set it to descriptor.
+    if (axisBufferPtr != nullptr)
+    {
+        for (uint32_t i = 0; i < inputTensorInfo1.GetNumElements(); ++i)
+        {
+            desc.m_vAxis.push_back(armnnUtils::GetUnsignedAxis(inputTensorInfo0.GetNumDimensions(),
+                                                               axisBufferPtr->data.data()[i]));
+        }
+    }
+
+    desc.m_TargetHeight    = input0Shape[1];
+    desc.m_TargetWidth     = input0Shape[2];
+    desc.m_KeepDims        = options->keep_dims;
+    desc.m_ReduceOperation = armnn::ReduceOperation::Sum;
+
+    // Register a new layer object, Sum.
+    IConnectableLayer *layer = m_Network->AddReduceLayer(desc, layerName.c_str());
+
+    armnn::TensorInfo outputTensorInfo = ToTensorInfo(outputs[0]);
+    layer->GetOutputSlot(0).SetTensorInfo(outputTensorInfo);
+
+    // Register input tensor to the layer.
+    auto inputTensorIndexes = AsUnsignedVector(GetInputTensorIds(m_Model, subgraphIndex, operatorIndex));
+    RegisterInputSlots(subgraphIndex, operatorIndex, layer, {inputTensorIndexes[0]});
+
+    // Register output tensor to the layer.
+    auto outputTensorIndexes = AsUnsignedVector(GetOutputTensorIds(m_Model, subgraphIndex, operatorIndex));
+    RegisterOutputSlots(subgraphIndex, operatorIndex, layer, outputTensorIndexes);
+}
+
 armnn::IConnectableLayer* TfLiteParser::AddFusedActivationLayer(armnn::IConnectableLayer* prevLayer,
                                                                 unsigned int outputSlot,
                                                                 tflite::ActivationFunctionType activationType)