CPP/functions/CPPSplit.h

/*
 * Copyright (c) 2020-2021 Arm Limited.
 *
 * SPDX-License-Identifier: MIT
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to
 * deal in the Software without restriction, including without limitation the
 * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
 * sell copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in all
 * copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */
#ifndef ARM_COMPUTE_CPP_SPLIT_H
#define ARM_COMPUTE_CPP_SPLIT_H

#include "arm_compute/core/Error.h"
#include "arm_compute/core/Helpers.h"
#include "arm_compute/core/TensorInfo.h"
#include "arm_compute/core/Types.h"
#include "arm_compute/core/utils/misc/ShapeCalculator.h"

#include "support/ToolchainSupport.h"

#include "arm_compute/runtime/IFunction.h"

namespace arm_compute
{
/** Basic function to split a tensor along a given axis */
template <typename SliceType, typename TensorInterfaceType = ITensor>
class CPPSplit : public IFunction
{
public:
    CPPSplit()
        : _outputs_vector(), _slice_functions(), _num_outputs(0)
    {
    }
    /** Static function to check if given info will lead to a valid configuration of @ref CPPSplit
     *
     * @param[in] input   The input tensor info. Data types supported: All.
     * @param[in] outputs A vector containing the output tensors' info. Data types supported: same as @p input.
     *                    The output tensors should match the input tensor dimensions for all shape dimensions apart
     *                    from the split dimension
     * @param[in] axis    Axis on which to split the input.
     *
     * @return a status
     */
    static Status validate(const ITensorInfo *input, const std::vector<ITensorInfo *> &outputs, unsigned int axis)
    {
        ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(input);
        ARM_COMPUTE_RETURN_ERROR_ON(axis >= input->num_dimensions());
        ARM_COMPUTE_RETURN_ERROR_ON(outputs.size() < 2);

        // Get output shape
        TensorShape  output_shape{};
        unsigned int total_output_shape_size = 0;

        // Sum the output sizes and fall back to evenly-sized splits if any are zero
        const bool using_split_shapes = std::none_of(outputs.begin(), outputs.end(), [&total_output_shape_size](ITensorInfo * info)
        {
            unsigned int output_shape_size = info->tensor_shape().total_size();
            total_output_shape_size += output_shape_size;
            return output_shape_size == 0;
        });

        if(using_split_shapes)
        {
            ARM_COMPUTE_RETURN_ERROR_ON(input->tensor_shape().total_size() != total_output_shape_size);
        }
        else
        {
            output_shape = arm_compute::misc::shape_calculator::compute_split_shape(input, axis, outputs.size());
            ARM_COMPUTE_RETURN_ERROR_ON(output_shape.total_size() == 0);
        }

        // Validate output tensors
        unsigned int axis_offset = 0;
        for(const auto &output : outputs)
        {
            ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(output);
            if(using_split_shapes)
            {
                output_shape = output->tensor_shape();
                ARM_COMPUTE_RETURN_ERROR_ON(output_shape.total_size() == 0);
            }

            const size_t axis_split_step = output_shape[axis];

            // Start/End coordinates
            Coordinates start_coords;
            Coordinates end_coords;
            for(unsigned int d = 0; d < output_shape.num_dimensions(); ++d)
            {
                end_coords.set(d, -1);
            }

            // Output auto inizialitation if not yet initialized
            TensorInfo tmp_output_info = *output->clone();
            if(tmp_output_info.tensor_shape().total_size() == 0)
            {
                tmp_output_info = input->clone()->set_is_resizable(true).set_tensor_shape(output_shape);
            }

            // Update coordinate on axis
            start_coords.set(axis, axis_offset);
            end_coords.set(axis, axis_offset + axis_split_step);

            ARM_COMPUTE_RETURN_ON_ERROR(SliceType::validate(input, output, start_coords, end_coords));
            axis_offset += axis_split_step;
        }

        return Status{};
    }

    /** Initialise the kernel's input and outputs.
     *
     * @param[in]  input   The input tensor. Data types supported: All
     * @param[out] outputs A vector containing the output tensors. Data types supported: Same as @p input.
     *                     The output tensors should match the input tensor dimensions for all shape dimensions apart
     *                     from the split dimension.
     * @param[in]  axis    Axis on which to split the input.
     */
    void configure(const TensorInterfaceType *input, const std::vector<TensorInterfaceType *> &outputs, unsigned int axis)
    {
        // Create Slice functions
        _num_outputs = outputs.size();
        _slice_functions.resize(_num_outputs);

        // Extract output tensor info
        std::vector<ITensorInfo *> outputs_info;
        for(auto &output : outputs)
        {
            ARM_COMPUTE_ERROR_ON_NULLPTR(output);
            outputs_info.emplace_back(output->info());
        }

        // If any of the outputs have a zero size, fall-back to using evenly-sized output splits
        const bool outputs_have_sizes = std::none_of(outputs_info.begin(), outputs_info.end(), [](ITensorInfo * info)
        {
            return info->tensor_shape().total_size() == 0;
        });

        // Validate
        ARM_COMPUTE_ERROR_THROW_ON(CPPSplit::validate(input->info(), outputs_info, axis));

        unsigned int axis_offset = 0;
        unsigned int i           = 0;

        for(const auto &output_info : outputs_info)
        {
            // Get output shape
            TensorShape output_shape = (outputs_have_sizes ?
                                        output_info->tensor_shape() :
                                        arm_compute::misc::shape_calculator::compute_split_shape(input->info(), axis, _num_outputs));

            const size_t axis_split_step = output_shape[axis];

            // Start/End coordinates
            Coordinates start_coords;
            Coordinates end_coords;

            for(unsigned int d = 0; d < output_shape.num_dimensions(); ++d)
            {
                end_coords.set(d, -1);
            }

            // Update coordinate on axis
            start_coords.set(axis, axis_offset);
            end_coords.set(axis, axis_offset + axis_split_step);

            // Configure slice function
            _slice_functions[i].configure(input, outputs[i], start_coords, end_coords);

            // Set valid region from shape
            outputs[i]->info()->set_valid_region(ValidRegion(Coordinates(), output_shape));

            // Update axis offset
            axis_offset += axis_split_step;
            ++i;
        }
    }

protected:
    std::vector<TensorInterfaceType *> _outputs_vector;
    std::vector<SliceType>             _slice_functions;
    unsigned int                       _num_outputs;
};

} // namespace arm_compute
#endif /* ARM_COMPUTE_CPP_SPLIT_H */