BnInference.cpp

#include <iostream>

#include <hipdnn_data_sdk/utilities/Tensor.hpp>
#include <hipdnn_frontend.hpp>

#include "hipdnn_data_sdk/utilities/Workspace.hpp"
#include "utils.hpp"

int main()
{
    using InputType = hipdnn_data_sdk::types::half;

    const int64_t n = 16; // Batch size
    // Input
    const int64_t c = 16; // Number of channels
    const int64_t h = 16; // Height
    const int64_t w = 16; // Width

    auto buildBnInferenceGraph = [=](hipdnnHandle_t handle) {
        auto graph = std::make_shared<hipdnn_frontend::graph::Graph>();
        graph->set_name("bn_inference_graph")
            .set_io_data_type(hipdnn_frontend::getDataTypeEnumFromType<InputType>())
            .set_intermediate_data_type(hipdnn_frontend::getDataTypeEnumFromType<InputType>())
            .set_compute_data_type(hipdnn_frontend::DataType::FLOAT);

        auto x = std::make_shared<hipdnn_frontend::graph::TensorAttributes>(
            hipdnn_frontend::graph::Tensor_attributes()
                .set_name("x")
                .set_dim({n, c, h, w})
                .set_stride({c * h * w, 1, c * w, c}));

        auto scale = std::make_shared<hipdnn_frontend::graph::TensorAttributes>(
            hipdnn_frontend::graph::Tensor_attributes()
                .set_name("scale")
                .set_dim({1, c, 1, 1})
                .set_stride({c, 1, c, c}));

        auto bias = std::make_shared<hipdnn_frontend::graph::TensorAttributes>(
            hipdnn_frontend::graph::Tensor_attributes()
                .set_name("bias")
                .set_dim({1, c, 1, 1})
                .set_stride({c, 1, c, c}));

        auto mean = std::make_shared<hipdnn_frontend::graph::TensorAttributes>(
            hipdnn_frontend::graph::Tensor_attributes()
                .set_name("mean")
                .set_dim({1, c, 1, 1})
                .set_stride({c, 1, c, c}));

        auto variance = std::make_shared<hipdnn_frontend::graph::TensorAttributes>(
            hipdnn_frontend::graph::Tensor_attributes()
                .set_name("variance")
                .set_dim({1, c, 1, 1})
                .set_stride({c, 1, c, c}));

        auto epsilon = std::make_shared<hipdnn_frontend::graph::TensorAttributes>(
            hipdnn_frontend::graph::Tensor_attributes()
                .set_name("epsilon")
                .set_dim({1, 1, 1, 1})
                .set_stride({1, 1, 1, 1})
                .set_value(1e-5));

        auto bnInferenceAttributes
            = hipdnn_frontend::graph::BatchnormInferenceAttributesVarianceExt().set_name(
                "bn_inference_node");

        auto y = graph->batchnorm_inference_variance_ext(
            x, mean, variance, scale, bias, epsilon, bnInferenceAttributes);
        y->set_output(true);

        // build graph
        HIPDNN_FE_CHECK(graph->build(handle));

        return std::make_tuple(graph, x, scale, bias, mean, variance, epsilon, y);
    };

    auto backend = hipdnn_frontend::detail::hipdnnBackend();
    if(!backend)
    {
        std::cout << "Creat backend failed. \n";
        return 1;
    }

    hipdnnHandle_t handle;
    HIPDNN_CHECK(backend->create(&handle));

    auto [graph, x, scale, bias, mean, variance, epsilon, y] = buildBnInferenceGraph(handle);

    hipdnn_data_sdk::utilities::Tensor<InputType> xTensor(x->get_dim(), x->get_stride());
    hipdnn_data_sdk::utilities::Tensor<InputType> scaleTensor(scale->get_dim());
    hipdnn_data_sdk::utilities::Tensor<InputType> biasTensor(bias->get_dim());
    hipdnn_data_sdk::utilities::Tensor<InputType> meanTensor(mean->get_dim());
    hipdnn_data_sdk::utilities::Tensor<InputType> varianceTensor(variance->get_dim());
    hipdnn_data_sdk::utilities::Tensor<InputType> epsilonTensor(epsilon->get_dim());
    hipdnn_data_sdk::utilities::Tensor<InputType> yTensor(y->get_dim(), y->get_stride());

    std::unordered_map<int64_t, void*> variantPack;
    variantPack[x->get_uid()] = xTensor.memory().deviceData();
    variantPack[scale->get_uid()] = scaleTensor.memory().deviceData();
    variantPack[bias->get_uid()] = biasTensor.memory().deviceData();
    variantPack[mean->get_uid()] = meanTensor.memory().deviceData();
    variantPack[variance->get_uid()] = varianceTensor.memory().deviceData();
    variantPack[epsilon->get_uid()] = epsilonTensor.memory().deviceData();
    variantPack[y->get_uid()] = yTensor.memory().deviceData();

    int64_t workspaceSize = 0;
    HIPDNN_FE_CHECK(graph->get_workspace_size(workspaceSize));
    const hipdnn_data_sdk::utilities::Workspace workspace(static_cast<size_t>(workspaceSize));

    HIPDNN_FE_CHECK(graph->execute(handle, variantPack, workspace.get()));

    std::cout << "Batch normalization inference graph execution complete. ";

    HIPDNN_CHECK(backend->destroy(handle));
    return 0;
}