模型训练测试之三：yolov5 模型训练及windows部署（二）

1、模型转换

模型需要在端侧进行部署时，需要对模型文件进行转换，官方项目中给了指定的转换方式export.py

--weights 指定转换模型文件路径，如 run\train\exp9\best.pt
--img-size 推理图片大小，根据指定的预训练模型对应的size大小
--batch-size 默认为1
--device 部署环境指定，GPU或者CPU
--include 部署类型，torchscript、onnx、coreml，c++版本一般需要第一个torchscript


python export.py --weights G:\Cxxx_Code\yolov5\runs\train\exp\weights\best.pt  --device cpu

得到best.torchscript.pt文件，之后就可以进行c++部署。

2、windows端VisualStudio部署

#include <opencv2/opencv.hpp>
#include <torch/script.h>
#include <torch/torch.h>
#include <algorithm>
#include <iostream>
#include <time.h>
#include <vector>
#include <memory>
using namespace cv;

std::vector<torch::Tensor> non_max_suppression(torch::Tensor preds, float score_thresh = 0.5, float iou_thresh = 0.5)
{
	std::vector<torch::Tensor> output;
	for (size_t i = 0; i < preds.sizes()[0]; ++i)
	{
		torch::Tensor pred = preds.select(0, i);

		// Filter by scores
		torch::Tensor scores = pred.select(1, 4) * std::get<0>(torch::max(pred.slice(1, 5, pred.sizes()[1]), 1));
		pred = torch::index_select(pred, 0, torch::nonzero(scores > score_thresh).select(1, 0));
		if (pred.sizes()[0] == 0) continue;

		// (center_x, center_y, w, h) to (left, top, right, bottom)
		pred.select(1, 0) = pred.select(1, 0) - pred.select(1, 2) / 2;
		pred.select(1, 1) = pred.select(1, 1) - pred.select(1, 3) / 2;
		pred.select(1, 2) = pred.select(1, 0) + pred.select(1, 2);
		pred.select(1, 3) = pred.select(1, 1) + pred.select(1, 3);

		// Computing scores and classes
		std::tuple<torch::Tensor, torch::Tensor> max_tuple = torch::max(pred.slice(1, 5, pred.sizes()[1]), 1);
		pred.select(1, 4) = pred.select(1, 4) * std::get<0>(max_tuple);
		pred.select(1, 5) = std::get<1>(max_tuple);

		torch::Tensor  dets = pred.slice(1, 0, 6);

		torch::Tensor keep = torch::empty({ dets.sizes()[0] });
		torch::Tensor areas = (dets.select(1, 3) - dets.select(1, 1)) * (dets.select(1, 2) - dets.select(1, 0));
		std::tuple<torch::Tensor, torch::Tensor> indexes_tuple = torch::sort(dets.select(1, 4), 0, 1);
		torch::Tensor v = std::get<0>(indexes_tuple);
		torch::Tensor indexes = std::get<1>(indexes_tuple);
		int count = 0;
		while (indexes.sizes()[0] > 0)
		{
			keep[count] = (indexes[0].item().toInt());
			count += 1;

			// Computing overlaps
			torch::Tensor lefts = torch::empty(indexes.sizes()[0] - 1);
			torch::Tensor tops = torch::empty(indexes.sizes()[0] - 1);
			torch::Tensor rights = torch::empty(indexes.sizes()[0] - 1);
			torch::Tensor bottoms = torch::empty(indexes.sizes()[0] - 1);
			torch::Tensor widths = torch::empty(indexes.sizes()[0] - 1);
			torch::Tensor heights = torch::empty(indexes.sizes()[0] - 1);
			for (size_t i = 0; i < indexes.sizes()[0] - 1; ++i)
			{
				lefts[i] = std::max(dets[indexes[0]][0].item().toFloat(), dets[indexes[i + 1]][0].item().toFloat());
				tops[i] = std::max(dets[indexes[0]][1].item().toFloat(), dets[indexes[i + 1]][1].item().toFloat());
				rights[i] = std::min(dets[indexes[0]][2].item().toFloat(), dets[indexes[i + 1]][2].item().toFloat());
				bottoms[i] = std::min(dets[indexes[0]][3].item().toFloat(), dets[indexes[i + 1]][3].item().toFloat());
				widths[i] = std::max(float(0), rights[i].item().toFloat() - lefts[i].item().toFloat());
				heights[i] = std::max(float(0), bottoms[i].item().toFloat() - tops[i].item().toFloat());
			}
			torch::Tensor overlaps = widths * heights;

			// FIlter by IOUs
			torch::Tensor ious = overlaps / (areas.select(0, indexes[0].item().toInt()) + torch::index_select(areas, 0, indexes.slice(0, 1, indexes.sizes()[0])) - overlaps);
			indexes = torch::index_select(indexes, 0, torch::nonzero(ious <= iou_thresh).select(1, 0) + 1);
		}
		keep = keep.toType(torch::kInt64);
		output.push_back(torch::index_select(dets, 0, keep.slice(0, 0, count)));
	}
	return output;
}


int main()
{
    // 测试cuda是否可用
	std::cout << "cuda::is_available():--------- " << torch::cuda::is_available() << "\n" << std::endl;
	torch::DeviceType device_type = at::kCPU; // 定义设备类型
	if (torch::cuda::is_available())
		device_type = at::kCUDA;
	// 加载模型
	torch::jit::script::Module module = torch::jit::load("best.torchscript.pt");
	module.to(device_type); // 模型加载至CPU
	//加载标签文件coco.names
	std::vector<std::string> classnames;
	std::ifstream f("coco.names");
	std::string name = "";
	while (std::getline(f, name))
	{
		std::cout << "name: " << name << std::endl;
		classnames.push_back(name);
	}
    
	int image_ind = 0;
	// 读取测试图片
    std::string image_file = "G:/xxxxx/test.jpg";
	Mat src = imread(image_file);
	if (src.empty())
	{
		printf("could not load image...\n");
		return -1;
	}
    // 计时器
	clock_t start = clock();
	int width = src.size().width;
	int height = src.size().height;
    // resize图片，和转换py中配置的参数一致即可
	cv::Mat img;
	resize(src, img, cv::Size(640, 640));
	cv::cvtColor(img, img, cv::COLOR_BGR2RGB);
	//输入数据移到cpu上
	torch::Tensor img_tensor = torch::from_blob(img.data, { 1, img.rows, img.cols, 3 }, torch::kByte).to(torch::kCPU);
	img_tensor = img_tensor.permute({ 0, 3, 1, 2 });
	img_tensor = img_tensor.toType(torch::kFloat);
	img_tensor = img_tensor.div(255);
	//模型推理
	std::vector<torch::jit::IValue> inputs;
	inputs.push_back(img_tensor);
	torch::Tensor preds = module.forward({ inputs }).toTuple()->elements()[0].toTensor();
    //nms处理
	std::vector<torch::Tensor> dets = non_max_suppression(preds, 0.5, 0.25);
	//打印结果，并可视化
	std::vector<std::string> result_all;
	for (size_t i = 0; i < dets[0].sizes()[0]; ++i)
	{
		float left = dets[0][i][0].item().toFloat() * src.cols / 640;
		float top = dets[0][i][1].item().toFloat() * src.rows / 640;
		float right = dets[0][i][2].item().toFloat() * src.cols / 640;
		float bottom = dets[0][i][3].item().toFloat() * src.rows / 640;
		float width = right - left;
		float height = bottom - top;
		float center_x = left + (width / 2);
		float center_y = top + (height / 2);
		float score = dets[0][i][4].item().toFloat();
		int classID = dets[0][i][5].item().toInt();

		std::cout << classnames[classID] << ":" << left << ", " << top << ", " << right << ", " << bottom << "\n" << std::endl;
		//在图片绘制结果
		cv::rectangle(src, cv::Rect(left, top, (right - left), (bottom - top)), cv::Scalar(0, 0, 255), 1);
		cv::putText(src, classnames[classID] + ": " + cv::format("%.2f", score), cv::Point(left, top), cv::FONT_HERSHEY_SIMPLEX, (right - left) / 200, cv::Scalar(0, 0, 255), 2);
	}

	cv::putText(src, "FPS: " + std::to_string(int(1e7 / (clock() - start))), cv::Point(50, 50), cv::FONT_HERSHEY_SIMPLEX, 1, cv::Scalar(0, 255, 0), 3);
	cv::namedWindow("src", cv::WINDOW_NORMAL);
	cv::imshow("src", src);
	cv::imwrite("predict.jpg", src);
	cv::waitKey(0);
}

配置下torch库（需要和自己训练的torch版本一致），opencv库；
结果如下