Merge branch 'Add_Shape_and_Split' into 'dev'

Add Shape and Split operators

See merge request !134

include/aidge/aidge.hpp

@@ -59,9 +59,11 @@
 #include "aidge/operator/ReduceMean.hpp"
 #include "aidge/operator/ReLU.hpp"
 #include "aidge/operator/Reshape.hpp"
+#include "aidge/operator/Shape.hpp"
 #include "aidge/operator/Scaling.hpp"
 #include "aidge/operator/Slice.hpp"
 #include "aidge/operator/Softmax.hpp"
+#include "aidge/operator/Split.hpp"
 #include "aidge/operator/Sqrt.hpp"
 #include "aidge/operator/Sub.hpp"
 #include "aidge/operator/Transpose.hpp"

include/aidge/operator/Shape.hpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#ifndef AIDGE_CORE_OPERATOR_SHAPE_H_
+#define AIDGE_CORE_OPERATOR_SHAPE_H_
+
+#include <cstdint>  // std::int64_t
+#include <memory>
+#include <string>
+#include <vector>
+
+#include "aidge/backend/OperatorImpl.hpp"
+#include "aidge/graph/Node.hpp"
+#include "aidge/operator/OperatorTensor.hpp"
+#include "aidge/utils/Registrar.hpp"
+#include "aidge/utils/StaticAttributes.hpp"
+#include "aidge/utils/Types.h"
+
+namespace Aidge {
+class Shape_OpImpl : public OperatorImpl {
+public:
+    Shape_OpImpl(const Operator& op, const std::string& backend = ""): OperatorImpl(op, backend) {}
+    void forward() override;
+};
+
+enum class ShapeAttr { Start, End };
+
+class Shape_Op : public OperatorTensor,
+                public Registrable<Shape_Op,
+                                   std::string,
+                                   std::shared_ptr<OperatorImpl>(const Shape_Op&)>,
+                public StaticAttributes<ShapeAttr, std::int64_t, std::int64_t> {
+
+public:
+    static const std::string Type;
+
+    Shape_Op() = delete;
+
+    using Attributes_ = StaticAttributes<ShapeAttr, std::int64_t, std::int64_t>;
+    template <ShapeAttr e> using attr = typename Attributes_::template attr<e>;
+    Shape_Op(std::int64_t start, std::int64_t end)
+            : OperatorTensor(Type, 1, 0, 1),
+            Attributes_(attr<ShapeAttr::Start>(start),
+                        attr<ShapeAttr::End>(end))
+    {
+        mImpl = std::make_shared<Shape_OpImpl>(*this);
+    }
+
+    /**
+     * @brief Copy-constructor. Copy the operator attributes and its output tensor(s), but not its input tensors (the new operator has no input associated).
+     * @param op Operator to copy.
+     */
+    Shape_Op(const Shape_Op& op)
+        : OperatorTensor(op),
+          Attributes_(op)
+    {
+        if (!op.backend().empty()) {
+            SET_IMPL_MACRO(Shape_Op, *this, op.backend());
+        }
+        else {
+            mImpl = std::make_shared<Shape_OpImpl>(*this);
+        }
+    }
+
+    /**
+     * @brief Clone the operator using its copy-constructor.
+     * @see Operator::Shape_Op
+     */
+    std::shared_ptr<Operator> clone() const override {
+        return std::make_shared<Shape_Op>(*this);
+    }
+
+    bool forwardDims(bool /*allowDataDependency*/ = false) override final;
+
+    void setBackend(const std::string& name, DeviceIdx_t device = 0) override;
+
+    static const std::vector<std::string> getInputsName(){
+        return {"data_input"};
+    }
+    static const std::vector<std::string> getOutputsName(){
+        return {"data_output"};
+    }
+};
+
+inline std::shared_ptr<Node> Shape(std::int64_t start = 0, std::int64_t end = -1, const std::string& name = "") {
+    return std::make_shared<Node>(std::make_shared<Shape_Op>(start, end), name);
+}
+} // namespace Aidge
+
+namespace {
+template <>
+const char *const EnumStrings<Aidge::ShapeAttr>::data[] = {"Start", "End"};
+}
+
+#endif /* AIDGE_CORE_OPERATOR_SHAPE_H_ */

include/aidge/operator/Split.hpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#ifndef AIDGE_CORE_OPERATOR_SPLIT_H_
+#define AIDGE_CORE_OPERATOR_SPLIT_H_
+
+#include <memory>
+#include <vector>
+
+#include "aidge/backend/OperatorImpl.hpp"
+#include "aidge/data/Tensor.hpp"
+#include "aidge/graph/Node.hpp"
+#include "aidge/operator/OperatorTensor.hpp"
+#include "aidge/utils/Registrar.hpp"
+#include "aidge/utils/StaticAttributes.hpp"
+#include "aidge/utils/Types.h"
+
+namespace Aidge {
+class Split_OpImpl : public OperatorImpl {
+public:
+    Split_OpImpl(const Operator& op, const std::string& backend = ""): OperatorImpl(op, backend) {}
+    void forward() override;
+};
+
+enum class SplitAttr { Axis, Split };
+
+class Split_Op
+    : public OperatorTensor,
+      public Registrable<Split_Op, std::string, std::shared_ptr<OperatorImpl>(const Split_Op &)>,
+      public StaticAttributes<SplitAttr, std::int8_t, std::vector<DimSize_t>> {
+
+public:
+    static const std::string Type;
+
+    Split_Op() = delete;
+
+    using Attributes_ = StaticAttributes<SplitAttr,  std::int8_t, std::vector<DimSize_t>>;
+    template <SplitAttr e> using attr = typename Attributes_::template attr<e>;
+    Split_Op( std::int8_t axis, DimSize_t nbOutputs, const std::vector<DimSize_t>& split)
+        : OperatorTensor(Type, 2, 0, nbOutputs),
+          Attributes_(attr<SplitAttr::Axis>(axis),
+                      attr<SplitAttr::Split>(split))
+    {
+        mImpl = std::make_shared<Split_OpImpl>(*this);
+    }
+
+
+    /**
+     * @brief Copy-constructor. Copy the operator attributes and its output tensor(s), but not its
+     * input tensors (the new operator has no input associated).
+     * @param op Operator to copy.
+     */
+    Split_Op(const Split_Op &op)
+        : OperatorTensor(op),
+          Attributes_(op)
+    {
+        if (!op.backend().empty()) {
+            SET_IMPL_MACRO(Split_Op, *this, op.backend());
+        }
+        else {
+            mImpl = std::make_shared<Split_OpImpl>(*this);
+        }
+    }
+public:
+    /**
+     * @brief Clone the operator using its copy-constructor.
+     * @see Operator::Split_Op
+     */
+    std::shared_ptr<Operator> clone() const override { return std::make_shared<Split_Op>(*this); }
+
+    bool dimsForwarded() const override final;
+    bool forwardDims(bool allowDataDependency = false) override final;
+
+    void setBackend(const std::string &name, DeviceIdx_t device = 0) override;
+
+    static const std::vector<std::string> getInputsName(){
+        return {"data_input", "split"};
+    }
+    static const std::vector<std::string> getOutputsName(){
+        return {"data_output_0", "data_output_n"};
+    }
+
+};
+
+/**
+ * @brief Exract a sub-Tensor from a bigger original Tensor.
+ * @param name Name of the Operator.
+ * @return std::shared_ptr<Node> A Node containing the Operator.
+ */
+inline std::shared_ptr<Node> Split(DimSize_t nbOutput,
+                                   std::int8_t axis = 0,
+                                   const std::vector<DimSize_t>& split = {},
+                                   const std::string &name = "") {
+    return std::make_shared<Node>(std::make_shared<Split_Op>(axis, nbOutput, split), name);
+}
+}  // namespace Aidge
+
+namespace {
+template <>
+const char *const EnumStrings<Aidge::SplitAttr>::data[] = { "Axis", "Split" };
+}
+
+#endif /* AIDGE_CORE_OPERATOR_SPLIT_H_ */

python_binding/operator/pybind_Shape.cpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#include <pybind11/pybind11.h>
+#include <string>
+#include <vector>
+
+#include "aidge/data/Tensor.hpp"
+#include "aidge/operator/Shape.hpp"
+#include "aidge/operator/OperatorTensor.hpp"
+
+namespace py = pybind11;
+namespace Aidge {
+
+void init_Shape(py::module& m) {
+    py::class_<Shape_Op, std::shared_ptr<Shape_Op>, Attributes, OperatorTensor>(m, "ShapeOp", py::multiple_inheritance())
+        .def(py::init<std::int64_t,
+                      std::int64_t>(),
+                py::arg("start"),
+                py::arg("end"))
+        .def_static("get_inputs_name", &Shape_Op::getInputsName)
+        .def_static("get_outputs_name", &Shape_Op::getOutputsName)
+        .def_static("attributes_name", &Shape_Op::staticGetAttrsName);
+
+    declare_registrable<Shape_Op>(m, "ShapeOp");
+
+    m.def("Shape", &Shape, py::arg("start") = 0, py::arg("end") = -1, py::arg("name") = "");
+}
+
+}  // namespace Aidge

python_binding/operator/pybind_Split.cpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#include <pybind11/pybind11.h>
+#include <string>
+#include <vector>
+
+#include "aidge/data/Tensor.hpp"
+#include "aidge/operator/Split.hpp"
+#include "aidge/operator/OperatorTensor.hpp"
+
+namespace py = pybind11;
+namespace Aidge {
+
+void init_Split(py::module& m) {
+    py::class_<Split_Op, std::shared_ptr<Split_Op>, Attributes, OperatorTensor>(m, "SplitOp", py::multiple_inheritance())
+        .def(py::init<DimSize_t, std::int8_t, std::vector<DimSize_t>&>(),
+                py::arg("nb_outputs"),
+                py::arg("axis"),
+                py::arg("split"))
+        .def_static("get_inputs_name", &Split_Op::getInputsName)
+        .def_static("get_outputs_name", &Split_Op::getOutputsName)
+        .def_static("attributes_name", &Split_Op::staticGetAttrsName);
+
+    declare_registrable<Split_Op>(m, "SplitOp");
+
+    m.def("Split", &Split, py::arg("nb_outputs"), py::arg("axis") = 0, py::arg("split") = std::vector<DimSize_t>(), py::arg("name") = "");
+}
+
+}  // namespace Aidge

python_binding/pybind_core.cpp

@@ -52,9 +52,11 @@ void init_ReduceMean(py::module&);
 void init_ReLU(py::module&);
 void init_Reshape(py::module&);
 void init_Scaling(py::module&);
+void init_Shape(py::module&);
 void init_Sigmoid(py::module&);
 void init_Slice(py::module&);
 void init_Softmax(py::module&);
+void init_Split(py::module&);
 void init_Sqrt(py::module&);
 void init_Sub(py::module&);
 void init_Tanh(py::module&);
@@ -120,9 +122,11 @@ void init_Aidge(py::module& m) {
     init_ReLU(m);
     init_Reshape(m);
     init_Scaling(m);
+    init_Shape(m);
     init_Sigmoid(m);
     init_Slice(m);
     init_Softmax(m);
+    init_Split(m);
     init_Sqrt(m);
     init_Sub(m);
     init_Tanh(m);

src/operator/Shape.cpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#include <cstddef>  // std::size_t
+#include <cstdint>  // std::int64_t
+#include <string>
+#include <vector>
+
+#include "aidge/operator/Shape.hpp"
+#include "aidge/data/Tensor.hpp"
+#include "aidge/utils/ErrorHandling.hpp"
+#include "aidge/utils/Types.h"
+
+void Aidge::Shape_OpImpl::forward() {
+    const Shape_Op& op = dynamic_cast<const Shape_Op&>(mOp);
+    const auto start = op.template getAttr<std::int64_t>("Start");
+    const auto end = op.template getAttr<std::int64_t>("End");
+
+    op.getOutput(0)->getImpl()->copyCast(std::next(op.getInput(0)->dims().data(), 
+                                                   start),
+                                         DataType::UInt64,
+                                         end - start + 1);
+}
+
+const std::string Aidge::Shape_Op::Type = "Shape";
+
+bool Aidge::Shape_Op::forwardDims(bool /*allowDataDependency*/) {
+    // check data input has been associated
+    if (!getInput(0)) {
+        AIDGE_THROW_OR_ABORT(std::runtime_error, "{}: input #0 should be associated with a Tensor", type());
+    }
+
+    if (getInput(0)->empty()) {
+        return false;
+    }
+
+    if (this->template getAttr<std::int64_t>("Start") < 0)
+        this->template getAttr<std::int64_t>("Start") += static_cast<std::int64_t>(getInput(0)->nbDims());
+    if (this->template getAttr<std::int64_t>("End") < 0)
+        this->template getAttr<std::int64_t>("End") += static_cast<std::int64_t>(getInput(0)->nbDims());
+
+    const auto start = this->template getAttr<std::int64_t>("Start");
+    const auto end = this->template getAttr<std::int64_t>("End");
+    const auto nbDims = static_cast<std::int64_t>(getInput(0)->nbDims());
+    const DimSize_t roi = end - start + 1;
+
+    AIDGE_ASSERT(start < nbDims && end < nbDims, "'Start' and 'End' must be < {}", nbDims);
+    AIDGE_ASSERT(roi> 1, "Unvalid ROI for Shape");
+
+    mOutputs[0]->resize({roi});
+    return true;
+}
+
+void Aidge::Shape_Op::setBackend(const std::string& name, Aidge::DeviceIdx_t device) {
+    if (Registrar<Shape_Op>::exists({name})) {
+        SET_IMPL_MACRO(Shape_Op, *this, name);
+    }
+    else {
+        mImpl = std::make_shared<Shape_OpImpl>(*this);
+    }
+    mOutputs[0]->setBackend(name, device);
+}

src/operator/Split.cpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#include "aidge/operator/Split.hpp"
+
+#include <cassert>
+#include <cstddef>
+#include <cstdint>
+#include <string>
+#include <utility>
+#include <vector>
+
+#include <fmt/format.h>
+
+#include "aidge/backend/OperatorImpl.hpp"
+#include "aidge/data/Data.hpp"
+#include "aidge/data/Tensor.hpp"
+#include "aidge/utils/ErrorHandling.hpp"
+#include "aidge/utils/Types.h"
+
+void Aidge::Split_OpImpl::forward() {
+    const Split_Op& op = dynamic_cast<const Split_Op&>(mOp);
+    const auto axis = op.template getAttr<std::int8_t>("Axis");
+    const auto splits = op.template getAttr<std::vector<DimSize_t>>("Split");
+    const auto dims = op.getInput(0)->dims();
+
+    //Compute pre/post axis strides
+    const std::size_t stride_pre = std::accumulate(dims.cbegin(), dims.cbegin() + axis, 1, std::multiplies<std::size_t>());
+    const std::size_t stride_post = std::accumulate(dims.crbegin(), dims.crbegin() + dims.size() -1 - axis, 1, std::multiplies<std::size_t>());
+    for (auto i = 0; i < op.nbOutputs(); ++i)
+    {
+        DimIdx_t chunkIdxOnAxis = std::accumulate(splits.cbegin(), splits.cbegin() + i, 0) * stride_post;
+        DimIdx_t offset = 0;
+        for (std::size_t j = 0; j < stride_pre; ++j)
+        {
+            // Compute chunk position in input tensor
+            DimIdx_t idx = j * stride_post * dims[axis] + chunkIdxOnAxis;
+            // Copy chunk in ouput
+            op.getOutput(i)->getImpl()->copy(op.getInput(0)->getImpl()->rawPtr(idx),
+                                             splits[i] * stride_post, offset);
+            offset += splits[i] * stride_post;
+        }
+
+    }
+}
+
+const std::string Aidge::Split_Op::Type = "Split";
+
+bool Aidge::Split_Op::dimsForwarded() const {
+    if ((getInput(1) && !getInput(1)->empty()))
+    {
+        // output dims are data dependent
+        return false;
+    }
+
+    return OperatorTensor::dimsForwarded();
+}
+
+bool Aidge::Split_Op::forwardDims(bool allowDataDependency) {
+    // check inputs have been associated
+    if (!getInput(0)) {
+        AIDGE_THROW_OR_ABORT(std::runtime_error, "{}: input #0 should be associated with a Tensor", type());
+    }
+
+    if (getInput(0)->empty()) {
+        return false;
+    }
+
+    std::shared_ptr<Tensor> fallback;
+
+    if (getInput(1) && !getInput(1)->empty()) { // Split is given, replace
+        if (!this->template getAttr<SplitAttr::Split>().empty()) {
+            Log::notice("Split_Op: ignoring non-empty Split attribute because input#1 takes precedence");
+        }
+
+        if (!allowDataDependency) {
+            Log::warn("Split_Op: unable to forwardDims() because output dims are data dependent on input#1");
+            return false;
+        }
+
+        this->template getAttr<SplitAttr::Split>().reserve(getInput(1)->size());
+        const auto& splits = getInput(1)->refCastFrom(fallback, NativeType<DimSize_t>::type, "cpu");
+        std::copy_n(static_cast<DimSize_t*>(splits.getImpl()->hostPtr()),
+                    splits.size(),
+                    std::back_inserter(this->template getAttr<SplitAttr::Split>()));
+    }
+
+    if (this->template getAttr<std::int8_t>("Axis") < 0)
+        this->template getAttr<std::int8_t>("Axis") += static_cast<std::int8_t>(getInput(0)->nbDims());
+
+    DimSize_t dimToSplit = getInput(0)->dims()[this->template getAttr<std::int8_t>("Axis")];
+    DimSize_t nbOutput = this->nbOutputs();
+    // Fill Split attr if empty
+    if(this->template getAttr<SplitAttr::Split>().empty()) {
+        // In case the input Split is not provided, divide the dimension of Axis into equal slices
+        AIDGE_ASSERT(dimToSplit > nbOutput, "Split_Op: Output number {} musn't be bigger than dimension {}.", nbOutput, dimToSplit);
+        DimSize_t baseSliceSize = dimToSplit / nbOutput;
+
+        DimSize_t remainder = dimToSplit % nbOutput;
+
+        for (DimSize_t i = 0; i < static_cast<DimSize_t>(nbOutput -1); ++i) {
+                this->template getAttr<SplitAttr::Split>().push_back(baseSliceSize);
+        }
+        this->template getAttr<SplitAttr::Split>().push_back(baseSliceSize + remainder);
+    }
+
+    const auto splits = this->template getAttr<SplitAttr::Split>();
+    AIDGE_ASSERT(splits.size() == nbOutput, "Split_Op: number of slices {} must be equal to number of outputs {}", splits, nbOutput);
+    DimSize_t totalSplitSize = std::accumulate(splits.cbegin(), splits.cend(), 0);
+    AIDGE_ASSERT(totalSplitSize == dimToSplit, "Split_Op: Total chunks size {} is different from dimension size {}.", totalSplitSize, dimToSplit);
+
+    std::vector<DimSize_t> outDims = getInput(0)->dims();
+    for (std::size_t i = 0; i < nbOutput; ++i)
+    {
+        outDims[this->template getAttr<std::int8_t>("Axis")] = this->template getAttr<SplitAttr::Split>()[i];
+        mOutputs[i]->resize(outDims);
+    }
+
+    return true;
+}
+
+void Aidge::Split_Op::setBackend(const std::string& name, Aidge::DeviceIdx_t device) {
+    if (Registrar<Split_Op>::exists({name})) {
+        SET_IMPL_MACRO(Split_Op, *this, name);
+    }
+    else {
+        mImpl = std::make_shared<Split_OpImpl>(*this);
+    }
+    for (std::size_t i = 0; i < this->nbOutputs(); i++)
+    {
+        mOutputs[i]->setBackend(name, device);
+    }
+    
+}

unit_tests/operator/Test_ShapeImpl.cpp

+/********************************************************************************
+ * Copyright (c) 2023 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#include <catch2/catch_test_macros.hpp>
+
+#include "aidge/data/Tensor.hpp"
+#include "aidge/operator/Shape.hpp"
+
+#include <cstdint>
+#include <memory>
+
+
+using namespace Aidge;
+
+TEST_CASE("[cpu/operator] Shape(forward)", "[Shape][CPU]") {
+    SECTION("Default attributes") {
+        std::shared_ptr<Tensor> input = std::make_shared<Tensor>(Array4D<int,1,2,3,5> {
+            {
+                {
+                    {
+                        { 1,  2,  3,  4,  5},
+                        { 6,  7,  8,  9, 10},
+                        {11, 12, 13, 14, 15}
+                    },
+                    {
+                        {16, 17, 18, 19, 20},
+                        {21, 22, 23, 24, 25},
+                        {26, 27, 28, 29, 30}
+                    }
+                }
+            }
+        });
+        std::shared_ptr<Tensor> expectedOutput = std::make_shared<Tensor>(Array1D<int,4> {
+            {1, 2, 3, 5}
+        });
+
+        std::shared_ptr<Node> myShape = Shape();
+        auto op = std::static_pointer_cast<OperatorTensor>(myShape -> getOperator());
+        op->associateInput(0,input);
+        op->setDataType(DataType::Int32);
+        op->setBackend("cpu");
+        myShape->forward();
+
+        REQUIRE(*(op->getOutput(0)) == *expectedOutput);
+
+    }
+    SECTION("Using attributes") {
+        std::shared_ptr<Tensor> input = std::make_shared<Tensor>(Array4D<int,1,2,3,5> {
+            {
+                {
+                    {
+                        { 1,  2,  3,  4,  5},
+                        { 6,  7,  8,  9, 10},
+                        {11, 12, 13, 14, 15}
+                    },
+                    {
+                        {16, 17, 18, 19, 20},
+                        {21, 22, 23, 24, 25},
+                        {26, 27, 28, 29, 30}
+                    }
+                }
+            }
+        });
+        std::shared_ptr<Tensor> expectedOutput = std::make_shared<Tensor>(Array1D<int,2> {
+            {2, 3}
+        });
+
+        std::shared_ptr<Node> myShape = Shape(1, 2);
+        auto op = std::static_pointer_cast<OperatorTensor>(myShape -> getOperator());
+        op->associateInput(0,input);
+        op->setDataType(DataType::Int32);
+        op->setBackend("cpu");
+        myShape->forward();
+
+        REQUIRE(*(op->getOutput(0)) == *expectedOutput);
+
+    }
+}
\ No newline at end of file

unit_tests/operator/Test_SplitImpl.cpp

+/********************************************************************************
+ * Copyright (c) 2024 CEA-List
+ *
+ * This program and the accompanying materials are made available under the
+ * terms of the Eclipse Public License 2.0 which is available at
+ * http://www.eclipse.org/legal/epl-2.0.
+ *
+ * SPDX-License-Identifier: EPL-2.0
+ *
+ ********************************************************************************/
+
+#include <catch2/catch_test_macros.hpp>
+
+#include "aidge/data/Tensor.hpp"
+#include "aidge/operator/Split.hpp"
+
+using namespace Aidge;
+
+TEST_CASE("[cpu/operator] Split(forward)", "[Split][CPU]") {
+    std::shared_ptr<Tensor> input = std::make_shared<Tensor>(Array4D<int,1,3,7,2> {
+        {
+            {
+                {{ 1,  2},{ 3,  4},{ 5,  6},{ 7,  8},{ 9, 10},{11, 12},{13, 14}},
+                {{15, 16},{17, 18},{19, 20},{21, 22},{23, 24},{25, 26},{27, 28}},
+                {{30, 31},{32, 33},{34, 35},{36, 37},{38, 39},{40, 41},{42, 43}}
+            }
+        }
+    });
+    SECTION("Default split") {
+        std::shared_ptr<Tensor> expectedOutput0 = std::make_shared<Tensor>(Array4D<int,1,3,2,2> {
+            {
+                {
+                    {{ 1,  2},{ 3,  4}},
+                    {{15, 16},{17, 18}},
+                    {{30, 31},{32, 33}}
+                }
+            }
+        });
+        std::shared_ptr<Tensor> expectedOutput1 = std::make_shared<Tensor>(Array4D<int,1,3,2,2> {
+            {
+                {
+                    {{ 5,  6},{ 7,  8}},
+                    {{19, 20},{21, 22}},
+                    {{34, 35},{36, 37}}
+                }
+            }
+        });
+        std::shared_ptr<Tensor> expectedOutput2 = std::make_shared<Tensor>(Array4D<int,1,3,3,2> {
+            {
+                {
+                    {{ 9, 10},{11, 12},{13, 14}},
+                    {{23, 24},{25, 26},{27, 28}},
+                    {{38, 39},{40, 41},{42, 43}}
+                }
+            }
+        });
+        auto mySplit = Split(DimSize_t(3), int8_t(2)); // Split on axis 2 into 3 outputs
+        mySplit->getOperator()->associateInput(0, input);
+        mySplit->getOperator()->setBackend("cpu");
+        mySplit->getOperator()->setDataType(DataType::Int32);
+        mySplit->forward();
+
+        REQUIRE(*std::static_pointer_cast<OperatorTensor>(mySplit->getOperator())->getOutput(0) == *expectedOutput0);
+        REQUIRE(*std::static_pointer_cast<OperatorTensor>(mySplit->getOperator())->getOutput(1) == *expectedOutput1);
+        REQUIRE(*std::static_pointer_cast<OperatorTensor>(mySplit->getOperator())->getOutput(2) == *expectedOutput2);
+    }
+    SECTION("Split with different chunk size") {
+        std::shared_ptr<Tensor> expectedOutput0 = std::make_shared<Tensor>(Array4D<int,1,3,4,2> {
+            {
+                {
+                    {{ 1,  2},{ 3,  4},{ 5,  6},{ 7,  8}},
+                    {{15, 16},{17, 18},{19, 20},{21, 22}},
+                    {{30, 31},{32, 33},{34, 35},{36, 37}}
+                }
+            }
+        });
+        std::shared_ptr<Tensor> expectedOutput1 = std::make_shared<Tensor>(Array4D<int,1,3,1,2> {
+            {
+                {
+                    {{ 9, 10}},
+                    {{23, 24}},
+                    {{38, 39}}
+                }
+            }
+        });
+        std::shared_ptr<Tensor> expectedOutput2 = std::make_shared<Tensor>(Array4D<int,1,3,2,2> {
+            {
+                {
+                    {{11, 12},{13, 14}},
+                    {{25, 26},{27, 28}},
+                    {{40, 41},{42, 43}}
+                }
+            }
+        });
+        auto mySplit = Split(DimSize_t(3), int8_t(2), {DimSize_t(4), DimSize_t(1), DimSize_t(2)});
+        mySplit->getOperator()->associateInput(0, input);
+        mySplit->getOperator()->setBackend("cpu");
+        mySplit->getOperator()->setDataType(DataType::Int32);
+        mySplit->forward();
+
+        REQUIRE(*std::static_pointer_cast<OperatorTensor>(mySplit->getOperator())->getOutput(0) == *expectedOutput0);
+        REQUIRE(*std::static_pointer_cast<OperatorTensor>(mySplit->getOperator())->getOutput(1) == *expectedOutput1);
+        REQUIRE(*std::static_pointer_cast<OperatorTensor>(mySplit->getOperator())->getOutput(2) == *expectedOutput2);
+    }
+    SECTION("Split with bad split attribute") {
+        auto mySplit = Split(DimSize_t(3), int8_t(2), {DimSize_t(4), DimSize_t(1), DimSize_t(3)});
+        mySplit->getOperator()->associateInput(0, input);
+        mySplit->getOperator()->setBackend("cpu");
+        mySplit->getOperator()->setDataType(DataType::Int32);
+        REQUIRE_THROWS(mySplit->forward());
+    }
+    SECTION("Split with bad outNumber") {
+        auto mySplit = Split(DimSize_t(8), int8_t(2));
+        mySplit->getOperator()->associateInput(0, input);
+        mySplit->getOperator()->setBackend("cpu");
+        mySplit->getOperator()->setDataType(DataType::Int32);
+        REQUIRE_THROWS(mySplit->forward());
+    }
+}
\ No newline at end of file