#include "InferenceModelEngine.h"

using namespace std;
using namespace ai_matrix;


InferenceModelEngine::InferenceModelEngine() {}
InferenceModelEngine::~InferenceModelEngine() {}


APP_ERROR InferenceModelEngine::Init()
{
    strPort0_ = engineName_ + "_" + std::to_string(engineId_) + "_0";

    //创建模型推理CUDA流   
    inference_model_stream_ = new cudaStream_t;
    CUDA_CHECK(cudaStreamCreate(inference_model_stream_));

    gLogger_ = new Logger;

    //相关资源分配
    buffers_[0] = nullptr; buffers_[1] = nullptr;
    CUDA_CHECK(cudaMalloc((void**)&buffers_[0], BATCH_SIZE * 3 * INPUT_H * INPUT_W * sizeof(float)));   //输入资源分配
    CUDA_CHECK(cudaMalloc((void**)&buffers_[1], BATCH_SIZE * OUTPUT_SIZE * sizeof(float)));    //输出资源分配

    LogInfo << "engineId_:" << engineId_ << " InferenceModelEngine Init ok";
    return APP_ERR_OK;
}

APP_ERROR InferenceModelEngine::DeInit()
{
    CUDA_CHECK(cudaStreamDestroy(*inference_model_stream_));  delete inference_model_stream_; inference_model_stream_ = nullptr; //释放模型推理CUDA流
    CUDA_CHECK(cudaFree(buffers_[0]));  //释放输入数据设备端内存
    CUDA_CHECK(cudaFree(buffers_[1])); //释放输出数据设备端内存

    //析构engine引擎资源
    context_->destroy();     //析构绘话
    engine_->destroy();      //析构TensorRT引擎
    runtime_->destroy();     //析构运行时环境

    delete gLogger_; gLogger_ = nullptr;

    LogInfo << "engineId_:" << engineId_ << " InferenceModelEngine DeInit ok";
    return APP_ERR_OK;
}


APP_ERROR InferenceModelEngine::Process()
{
    cudaSetDevice(DEVICE);  //设置GPU 

    //wts及engine模型名称
    std::string wts_name = MyYaml::GetIns()->GetStringValue("yolov5_wts_name");
    std::string engine_name = MyYaml::GetIns()->GetStringValue("yolov5_model_name");

    bool is_p6 = false; //默认不是P6模型

    /**********************************************************************************
        gw width_multiple系数: width_multiple控制网络的宽度。
        gd depth_multiple系数: depth_multiple控制网络的深度
        N模型:
            gd = 0.33;gw = 0.25;
        S模型:
            gd = 0.33;gw = 0.50;
        M模型:
            gd = 0.67;gw = 0.75;
        L模型:
            gd = 1.0;gw = 1.0;
        X模型:
            gd = 1.33;gw = 1.25;
     **********************************************************************************/
    float gd = 0.67, gw = 0.75;     //默认使用M模型


    //序列化引擎
    //直接使用API创建一个模型，并将其序列化为流  编译成TensorRT引擎engine文件后无需再次调用,调用依次生成engine即可
    #if 0
    if (!wts_name.empty()) {
        IHostMemory* modelStream{ nullptr };
        APIToModel(*gLogger_, BATCH_SIZE, &modelStream, is_p6, gd, gw, wts_name);
        assert(modelStream != nullptr);
        std::ofstream p(engine_name, std::ios::binary);
        if (!p) {
            std::cerr << "could not open plan output file" << std::endl;
            return -1;
        }
        p.write(reinterpret_cast<const char*>(modelStream->data()), modelStream->size());
        modelStream->destroy();
    }
    #endif

    //反序列化模型并运行推理
    std::ifstream file(engine_name, std::ios::binary);
    if (!file.good()) {
        LogInfo << "read " << engine_name << " error!" << std::endl;
        exit(0);
    }

    //创建tensorRT流对象trtModelStream,这个就跟文件流中的ifstream类似的
    //trtModelStream是一块内存区域,用于保存序列化的plan文件
    char *trtModelStream = nullptr;
    size_t size = 0;
    file.seekg(0, file.end);    //将指针移动至距离文件末尾0处的位置
    size = file.tellg();    //获得当前字符的位置
    file.seekg(0, file.beg);    //将指针移动至距离文件开头0处的位置
    trtModelStream = new char[size];
    assert(trtModelStream);
    file.read(trtModelStream, size);    //将序列化engine模型(数据及数据大小)读入trtModelStream
    file.close();

   
    //1.设置运行时环境
    //runtime_ = new IRuntime;
    runtime_ = createInferRuntime(*gLogger_);     //创建运行时环境IRuntime对象,传入gLogger用于打印信息
    assert(runtime_ != nullptr);
    //2.生成反序列化引擎
    //engine = new ICudaEngine;
    engine_ = runtime_->deserializeCudaEngine(trtModelStream, size); //反序列化引擎engine(根据trtModelStream反序列化)
    assert(engine_ != nullptr);
    //3.创建上下文环境
    //context = new IExecutionContext;
    context_ = engine_->createExecutionContext();  //创建上下文环境,主要用于inference函数中启动cuda核
    assert(context_ != nullptr);
    delete[] trtModelStream;    //析构trtModelStream
    assert(engine_->getNbBindings() == 2);
    
    //获取绑定的输入输入
    const int inputIndex = engine_->getBindingIndex(INPUT_BLOB_NAME);
    const int outputIndex = engine_->getBindingIndex(OUTPUT_BLOB_NAME);
    std::cout<<"inputIndex: "<<inputIndex<<"\toutputIndex: "<<outputIndex<<std::endl; 
    assert(inputIndex == 0);
    assert(outputIndex == 1);

    uint64_t u64count_num = 0;
    int iRet = APP_ERR_OK;
    
    while (!isStop_)
    {
        std::shared_ptr<void> pVoidData0 = nullptr;
        inputQueMap_[strPort0_]->pop(pVoidData0);
        if (nullptr == pVoidData0)
        {
            usleep(1*1000); //n ms
            continue;
        }
        // LogInfo << "receive from ImagePreprocessEngine's data success!";
        // std::cout<<"receive from ImagePreprocessEngine's data success!"<<std::endl;

        // std::cout<<"Enter InferenceModelEngine Thread "<<++u64count_num<<" Times!"<<std::endl;
        std::shared_ptr<InferenceData> pImagePreprocessData = std::static_pointer_cast<InferenceData>(pVoidData0);
        
        //将图像预处理数据拷贝到buffers_[0]
        #ifdef CUDA_MEMCPY_TIME_CONSUMING_TEST 
        auto cuda_memcpy_start = std::chrono::system_clock::now();  //计时开始
        CUDA_CHECK(cudaMemcpyAsync(buffers_[0], static_cast<void *>(pImagePreprocessData->pData.get()), pImagePreprocessData->iSize, cudaMemcpyDeviceToDevice,*inference_model_stream_));  
        auto cuda_memcpy_end = std::chrono::system_clock::now();  //计时结束
        std::cout<< "InferenceModelEngine cuda memcpy data size is: "<<pImagePreprocessData->iSize<<std::endl;
        std::cout << "InferenceModelEngine cuda memcpy device to device time: " << std::chrono::duration_cast<std::chrono::milliseconds>(cuda_memcpy_end - cuda_memcpy_start).count() << "ms" << std::endl;
        #else
        CUDA_CHECK(cudaMemcpyAsync(buffers_[0], static_cast<void *>(pImagePreprocessData->pData.get()), pImagePreprocessData->iSize, cudaMemcpyDeviceToDevice,*inference_model_stream_));  
        #endif

        //构造推理结果数据
        void* pInferenceModelBuffer = nullptr;
        unsigned int pInferenceModelBuffer_Size = BATCH_SIZE * OUTPUT_SIZE;
        pInferenceModelBuffer = new float[pInferenceModelBuffer_Size];

        void* pSrcRGBBuffer = nullptr;
		unsigned int pSrcRGBBuffer_Size = pImagePreprocessData->iSrcSize;
		pSrcRGBBuffer = new uint8_t[pSrcRGBBuffer_Size];
        memcpy(pSrcRGBBuffer, pImagePreprocessData->pSrcData.get(), pSrcRGBBuffer_Size);

        std::shared_ptr<InferenceData> pInferenceModelData = std::make_shared<InferenceData>();

        #ifdef INFERENCE_MODEL_TIME_CONSUMING_TEST 
        auto start = std::chrono::system_clock::now();  //计时开始
        doInference(*context_, *inference_model_stream_, (void**)buffers_, (float*)pInferenceModelBuffer, BATCH_SIZE);  //context为推理的上下文环境,stream为注册流(用于异步推理时进行同步),buffers为传入的图像数据,pInferenceModelBuffer为推理的结果
        auto end = std::chrono::system_clock::now();    
        std::cout << "inference time: " << std::chrono::duration_cast<std::chrono::milliseconds>(end - start).count() << "ms" << std::endl;
        #else
        doInference(*context_, *inference_model_stream_, (void**)buffers_, (float*)pInferenceModelBuffer, BATCH_SIZE);  //context为推理的上下文环境,stream为注册流(用于异步推理时进行同步),buffers为传入的图像数据,pInferenceModelBuffer为推理的结果
        #endif

        //组织数据
        pInferenceModelData->iDataSource = engineId_;
        pInferenceModelData->iSize = pInferenceModelBuffer_Size;
        pInferenceModelData->pData.reset(pInferenceModelBuffer, [](void* data){if(data){delete[] data; data = nullptr;}}); //智能指针管理内存
        pInferenceModelData->iSrcSize = pSrcRGBBuffer_Size;
        pInferenceModelData->pSrcData.reset(pSrcRGBBuffer, [](void* data){if(data){delete[] data; data = nullptr;}}); //智能指针管理内存
        pInferenceModelData->i64TimeStamp = pImagePreprocessData->i64TimeStamp;

        #if 1
        //推理结果送入下一引擎
        iRet = outputQueMap_[strPort0_]->push(std::static_pointer_cast<void>(pInferenceModelData));
        if (iRet != APP_ERR_OK){
			LogError << "push info error";
            // std::cerr<<"push the inference model data failed..."<<std::endl;
		}else{
            // std::cout<<"push the inference model data success!"<<std::endl;
        }
        #endif
    }
}



//获取宽度
int InferenceModelEngine::get_width(int x, float gw, int divisor = 8) {
    return int(ceil((x * gw) / divisor)) * divisor;
}

//获取深度
int InferenceModelEngine::get_depth(int x, float gd) {
    if (x == 1) return 1;
    int r = round(x * gd);
    if (x * gd - int(x * gd) == 0.5 && (int(x * gd) % 2) == 0) {
        --r;
    }
    return std::max<int>(r, 1);
}

//构建普通引擎(例:yolov5s,yolov5m...)
ICudaEngine* InferenceModelEngine::build_engine(unsigned int maxBatchSize, IBuilder* builder, IBuilderConfig* config, nvinfer1::DataType dt, float& gd, float& gw, std::string& wts_name) {
    INetworkDefinition* network = builder->createNetworkV2(0U);

    // Create input tensor of shape {3, INPUT_H, INPUT_W} with name INPUT_BLOB_NAME
    ITensor* data = network->addInput(INPUT_BLOB_NAME, dt, Dims3{ 3, INPUT_H, INPUT_W });
    assert(data);
    std::map<std::string, Weights> weightMap = loadWeights(wts_name);
    /* ------ yolov5 backbone------ */
    auto conv0 = convBlock(network, weightMap, *data,  get_width(64, gw), 6, 2, 1,  "model.0");
    assert(conv0);
    auto conv1 = convBlock(network, weightMap, *conv0->getOutput(0), get_width(128, gw), 3, 2, 1, "model.1");
    auto bottleneck_CSP2 = C3(network, weightMap, *conv1->getOutput(0), get_width(128, gw), get_width(128, gw), get_depth(3, gd), true, 1, 0.5, "model.2");
    auto conv3 = convBlock(network, weightMap, *bottleneck_CSP2->getOutput(0), get_width(256, gw), 3, 2, 1, "model.3");
    auto bottleneck_csp4 = C3(network, weightMap, *conv3->getOutput(0), get_width(256, gw), get_width(256, gw), get_depth(6, gd), true, 1, 0.5, "model.4");
    auto conv5 = convBlock(network, weightMap, *bottleneck_csp4->getOutput(0), get_width(512, gw), 3, 2, 1, "model.5");
    auto bottleneck_csp6 = C3(network, weightMap, *conv5->getOutput(0), get_width(512, gw), get_width(512, gw), get_depth(9, gd), true, 1, 0.5, "model.6");
    auto conv7 = convBlock(network, weightMap, *bottleneck_csp6->getOutput(0), get_width(1024, gw), 3, 2, 1, "model.7");
    auto bottleneck_csp8 = C3(network, weightMap, *conv7->getOutput(0), get_width(1024, gw), get_width(1024, gw), get_depth(3, gd), true, 1, 0.5, "model.8");
    auto spp9 = SPPF(network, weightMap, *bottleneck_csp8->getOutput(0), get_width(1024, gw), get_width(1024, gw), 5, "model.9");
    /* ------ yolov5 head ------ */
    auto conv10 = convBlock(network, weightMap, *spp9->getOutput(0), get_width(512, gw), 1, 1, 1, "model.10");

    auto upsample11 = network->addResize(*conv10->getOutput(0));
    assert(upsample11);
    upsample11->setResizeMode(ResizeMode::kNEAREST);
    upsample11->setOutputDimensions(bottleneck_csp6->getOutput(0)->getDimensions());

    ITensor* inputTensors12[] = { upsample11->getOutput(0), bottleneck_csp6->getOutput(0) };
    auto cat12 = network->addConcatenation(inputTensors12, 2);
    auto bottleneck_csp13 = C3(network, weightMap, *cat12->getOutput(0), get_width(1024, gw), get_width(512, gw), get_depth(3, gd), false, 1, 0.5, "model.13");
    auto conv14 = convBlock(network, weightMap, *bottleneck_csp13->getOutput(0), get_width(256, gw), 1, 1, 1, "model.14");

    auto upsample15 = network->addResize(*conv14->getOutput(0));
    assert(upsample15);
    upsample15->setResizeMode(ResizeMode::kNEAREST);
    upsample15->setOutputDimensions(bottleneck_csp4->getOutput(0)->getDimensions());

    ITensor* inputTensors16[] = { upsample15->getOutput(0), bottleneck_csp4->getOutput(0) };
    auto cat16 = network->addConcatenation(inputTensors16, 2);

    auto bottleneck_csp17 = C3(network, weightMap, *cat16->getOutput(0), get_width(512, gw), get_width(256, gw), get_depth(3, gd), false, 1, 0.5, "model.17");

    /* ------ detect ------ */
    IConvolutionLayer* det0 = network->addConvolutionNd(*bottleneck_csp17->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.24.m.0.weight"], weightMap["model.24.m.0.bias"]);
    auto conv18 = convBlock(network, weightMap, *bottleneck_csp17->getOutput(0), get_width(256, gw), 3, 2, 1, "model.18");
    ITensor* inputTensors19[] = { conv18->getOutput(0), conv14->getOutput(0) };
    auto cat19 = network->addConcatenation(inputTensors19, 2);
    auto bottleneck_csp20 = C3(network, weightMap, *cat19->getOutput(0), get_width(512, gw), get_width(512, gw), get_depth(3, gd), false, 1, 0.5, "model.20");
    IConvolutionLayer* det1 = network->addConvolutionNd(*bottleneck_csp20->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.24.m.1.weight"], weightMap["model.24.m.1.bias"]);
    auto conv21 = convBlock(network, weightMap, *bottleneck_csp20->getOutput(0), get_width(512, gw), 3, 2, 1, "model.21");
    ITensor* inputTensors22[] = { conv21->getOutput(0), conv10->getOutput(0) };
    auto cat22 = network->addConcatenation(inputTensors22, 2);
    auto bottleneck_csp23 = C3(network, weightMap, *cat22->getOutput(0), get_width(1024, gw), get_width(1024, gw), get_depth(3, gd), false, 1, 0.5, "model.23");
    IConvolutionLayer* det2 = network->addConvolutionNd(*bottleneck_csp23->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.24.m.2.weight"], weightMap["model.24.m.2.bias"]);

    auto yolo = addYoLoLayer(network, weightMap, "model.24", std::vector<IConvolutionLayer*>{det0, det1, det2});
    yolo->getOutput(0)->setName(OUTPUT_BLOB_NAME);
    network->markOutput(*yolo->getOutput(0));
    // Build engine
    builder->setMaxBatchSize(maxBatchSize);
    config->setMaxWorkspaceSize(16 * (1 << 20));  // 16MB
#if defined(USE_FP16)
    config->setFlag(BuilderFlag::kFP16);
#elif defined(USE_INT8)
    std::cout << "Your platform support int8: " << (builder->platformHasFastInt8() ? "true" : "false") << std::endl;
    assert(builder->platformHasFastInt8());
    config->setFlag(BuilderFlag::kINT8);
    Int8EntropyCalibrator2* calibrator = new Int8EntropyCalibrator2(1, INPUT_W, INPUT_H, "./coco_calib/", "int8calib.table", INPUT_BLOB_NAME);
    config->setInt8Calibrator(calibrator);
#endif

    std::cout << "Building engine, please wait for a while..." << std::endl;
    ICudaEngine* engine = builder->buildEngineWithConfig(*network, *config);
    std::cout << "Build engine successfully!" << std::endl;

    // Don't need the network any more
    network->destroy();

    // Release host memory
    for (auto& mem : weightMap)
    {
        free((void*)(mem.second.values));
    }

    return engine;
}

//构建p6引擎(例:yolov5s6,yolov5m6...)
ICudaEngine* InferenceModelEngine::build_engine_p6(unsigned int maxBatchSize, IBuilder* builder, IBuilderConfig* config, nvinfer1::DataType dt, float& gd, float& gw, std::string& wts_name) {
    INetworkDefinition* network = builder->createNetworkV2(0U);
    // Create input tensor of shape {3, INPUT_H, INPUT_W} with name INPUT_BLOB_NAME
    ITensor* data = network->addInput(INPUT_BLOB_NAME, dt, Dims3{ 3, INPUT_H, INPUT_W });
    assert(data);
    
    std::map<std::string, Weights> weightMap = loadWeights(wts_name);

    /* ------ yolov5 backbone------ */
    auto conv0 = convBlock(network, weightMap, *data,  get_width(64, gw), 6, 2, 1,  "model.0");
    auto conv1 = convBlock(network, weightMap, *conv0->getOutput(0), get_width(128, gw), 3, 2, 1, "model.1");
    auto c3_2 = C3(network, weightMap, *conv1->getOutput(0), get_width(128, gw), get_width(128, gw), get_depth(3, gd), true, 1, 0.5, "model.2");
    auto conv3 = convBlock(network, weightMap, *c3_2->getOutput(0), get_width(256, gw), 3, 2, 1, "model.3");
    auto c3_4 = C3(network, weightMap, *conv3->getOutput(0), get_width(256, gw), get_width(256, gw), get_depth(6, gd), true, 1, 0.5, "model.4");
    auto conv5 = convBlock(network, weightMap, *c3_4->getOutput(0), get_width(512, gw), 3, 2, 1, "model.5");
    auto c3_6 = C3(network, weightMap, *conv5->getOutput(0), get_width(512, gw), get_width(512, gw), get_depth(9, gd), true, 1, 0.5, "model.6");
    auto conv7 = convBlock(network, weightMap, *c3_6->getOutput(0), get_width(768, gw), 3, 2, 1, "model.7");
    auto c3_8 = C3(network, weightMap, *conv7->getOutput(0), get_width(768, gw), get_width(768, gw), get_depth(3, gd), true, 1, 0.5, "model.8");
    auto conv9 = convBlock(network, weightMap, *c3_8->getOutput(0), get_width(1024, gw), 3, 2, 1, "model.9");
    auto c3_10 = C3(network, weightMap, *conv9->getOutput(0), get_width(1024, gw), get_width(1024, gw), get_depth(3, gd), true, 1, 0.5, "model.10");
    auto sppf11 = SPPF(network, weightMap, *c3_10->getOutput(0), get_width(1024, gw), get_width(1024, gw), 5, "model.11");

    /* ------ yolov5 head ------ */
    auto conv12 = convBlock(network, weightMap, *sppf11->getOutput(0), get_width(768, gw), 1, 1, 1, "model.12");
    auto upsample13 = network->addResize(*conv12->getOutput(0));
    assert(upsample13);
    upsample13->setResizeMode(ResizeMode::kNEAREST);
    upsample13->setOutputDimensions(c3_8->getOutput(0)->getDimensions());
    ITensor* inputTensors14[] = { upsample13->getOutput(0), c3_8->getOutput(0) };
    auto cat14 = network->addConcatenation(inputTensors14, 2);
    auto c3_15 = C3(network, weightMap, *cat14->getOutput(0), get_width(1536, gw), get_width(768, gw), get_depth(3, gd), false, 1, 0.5, "model.15");

    auto conv16 = convBlock(network, weightMap, *c3_15->getOutput(0), get_width(512, gw), 1, 1, 1, "model.16");
    auto upsample17 = network->addResize(*conv16->getOutput(0));
    assert(upsample17);
    upsample17->setResizeMode(ResizeMode::kNEAREST);
    upsample17->setOutputDimensions(c3_6->getOutput(0)->getDimensions());
    ITensor* inputTensors18[] = { upsample17->getOutput(0), c3_6->getOutput(0) };
    auto cat18 = network->addConcatenation(inputTensors18, 2);
    auto c3_19 = C3(network, weightMap, *cat18->getOutput(0), get_width(1024, gw), get_width(512, gw), get_depth(3, gd), false, 1, 0.5, "model.19");

    auto conv20 = convBlock(network, weightMap, *c3_19->getOutput(0), get_width(256, gw), 1, 1, 1, "model.20");
    auto upsample21 = network->addResize(*conv20->getOutput(0));
    assert(upsample21);
    upsample21->setResizeMode(ResizeMode::kNEAREST);
    upsample21->setOutputDimensions(c3_4->getOutput(0)->getDimensions());
    ITensor* inputTensors21[] = { upsample21->getOutput(0), c3_4->getOutput(0) };
    auto cat22 = network->addConcatenation(inputTensors21, 2);
    auto c3_23 = C3(network, weightMap, *cat22->getOutput(0), get_width(512, gw), get_width(256, gw), get_depth(3, gd), false, 1, 0.5, "model.23");

    auto conv24 = convBlock(network, weightMap, *c3_23->getOutput(0), get_width(256, gw), 3, 2, 1, "model.24");
    ITensor* inputTensors25[] = { conv24->getOutput(0), conv20->getOutput(0) };
    auto cat25 = network->addConcatenation(inputTensors25, 2);
    auto c3_26 = C3(network, weightMap, *cat25->getOutput(0), get_width(1024, gw), get_width(512, gw), get_depth(3, gd), false, 1, 0.5, "model.26");

    auto conv27 = convBlock(network, weightMap, *c3_26->getOutput(0), get_width(512, gw), 3, 2, 1, "model.27");
    ITensor* inputTensors28[] = { conv27->getOutput(0), conv16->getOutput(0) };
    auto cat28 = network->addConcatenation(inputTensors28, 2);
    auto c3_29 = C3(network, weightMap, *cat28->getOutput(0), get_width(1536, gw), get_width(768, gw), get_depth(3, gd), false, 1, 0.5, "model.29");

    auto conv30 = convBlock(network, weightMap, *c3_29->getOutput(0), get_width(768, gw), 3, 2, 1, "model.30");
    ITensor* inputTensors31[] = { conv30->getOutput(0), conv12->getOutput(0) };
    auto cat31 = network->addConcatenation(inputTensors31, 2);
    auto c3_32 = C3(network, weightMap, *cat31->getOutput(0), get_width(2048, gw), get_width(1024, gw), get_depth(3, gd), false, 1, 0.5, "model.32");

    /* ------ detect ------ */
    IConvolutionLayer* det0 = network->addConvolutionNd(*c3_23->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.33.m.0.weight"], weightMap["model.33.m.0.bias"]);
    IConvolutionLayer* det1 = network->addConvolutionNd(*c3_26->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.33.m.1.weight"], weightMap["model.33.m.1.bias"]);
    IConvolutionLayer* det2 = network->addConvolutionNd(*c3_29->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.33.m.2.weight"], weightMap["model.33.m.2.bias"]);
    IConvolutionLayer* det3 = network->addConvolutionNd(*c3_32->getOutput(0), 3 * (Yolo::CLASS_NUM + 5), DimsHW{ 1, 1 }, weightMap["model.33.m.3.weight"], weightMap["model.33.m.3.bias"]);

    auto yolo = addYoLoLayer(network, weightMap, "model.33", std::vector<IConvolutionLayer*>{det0, det1, det2, det3});
    yolo->getOutput(0)->setName(OUTPUT_BLOB_NAME);
    network->markOutput(*yolo->getOutput(0));

    // Build engine
    builder->setMaxBatchSize(maxBatchSize);
    config->setMaxWorkspaceSize(16 * (1 << 20));  // 16MB
#if defined(USE_FP16)
    config->setFlag(BuilderFlag::kFP16);
#elif defined(USE_INT8)
    std::cout << "Your platform support int8: " << (builder->platformHasFastInt8() ? "true" : "false") << std::endl;
    assert(builder->platformHasFastInt8());
    config->setFlag(BuilderFlag::kINT8);
    Int8EntropyCalibrator2* calibrator = new Int8EntropyCalibrator2(1, INPUT_W, INPUT_H, "./coco_calib/", "int8calib.table", INPUT_BLOB_NAME);
    config->setInt8Calibrator(calibrator);
#endif

    std::cout << "Building engine, please wait for a while..." << std::endl;
    ICudaEngine* engine = builder->buildEngineWithConfig(*network, *config);
    std::cout << "Build engine successfully!" << std::endl;

    // Don't need the network any more
    network->destroy();

    // Release host memory
    for (auto& mem : weightMap)
    {
        free((void*)(mem.second.values));
    }

    return engine;
}

//转换模型
void InferenceModelEngine::APIToModel(Logger gLogger, unsigned int maxBatchSize, IHostMemory** modelStream, bool& is_p6, float& gd, float& gw, std::string& wts_name) {
    // Create builder
    IBuilder* builder = createInferBuilder(gLogger);    //创建builder(要传入gLogger)
    IBuilderConfig* config = builder->createBuilderConfig();    //创建builderconfig

    // 创建模型来填充网络，然后设置输出并创建一个引擎  
    // Create model to populate the network, then set the outputs and create an engine
    ICudaEngine *engine = nullptr;
    if (is_p6) {
        engine = build_engine_p6(maxBatchSize, builder, config, nvinfer1::DataType::kFLOAT, gd, gw, wts_name);
    } else {
        engine = build_engine(maxBatchSize, builder, config, nvinfer1::DataType::kFLOAT, gd, gw, wts_name);
    }
    assert(engine != nullptr);

    // Serialize the engine
    //序列化引擎生成模型流
    (*modelStream) = engine->serialize();

    // Close everything down
    //释放相关资源
    engine->destroy();
    builder->destroy();
    config->destroy();
}

//执行推理
void InferenceModelEngine::doInference(IExecutionContext& context, cudaStream_t& stream, void **buffers, float* output, int batchSize) {
    // infer on the batch asynchronously, and DMA output back to host
    context.enqueue(batchSize, buffers, stream, nullptr);   //执行异步推理(调用context->enqueueV2即可执行异步推理,如果用同步推理的话,可以调用context->executeV2)
    CUDA_CHECK(cudaMemcpyAsync(output, buffers[1], batchSize * OUTPUT_SIZE * sizeof(float), cudaMemcpyDeviceToHost, stream));   //把推理后的结果从GPU上拷贝到CPU上
    cudaStreamSynchronize(stream); //同步之前创建的cuda流，原因很简单，直接使用的context->enqueueV2函数是异步推理,因此需要把cuda流同步一下
}