使用yolov8的一些错误

出现这个报错的时候：

AutoInstall will run now for 'ultralytics.nn.modules.conv' but this feature will be removed in the future.
Recommend fixes are to train a new model using the latest 'ultralytics' package or to run a command with an official YOLOv8 model, i.e. 'yolo predict model=yolov8n.pt'
requirements: YOLOv8 requirement "ultralytics.nn.modules.conv" not found, attempting AutoUpdate...
ERROR: Could not find a version that satisfies the requirement ultralytics.nn.modules.conv (from versions: none)
ERROR: No matching distribution found for ultralytics.nn.modules.conv
requirements: ❌ Command 'pip install "ultralytics.nn.modules.conv"  ' returned non-zero exit status 1.File "/usr/local/ev_sdk/src/ultralytics/nn/tasks.py", line 351, in torch_safe_load
    return torch.load(file, map_location='cpu'), file  # load
  File "/opt/conda/lib/python3.7/site-packages/torch/serialization.py", line 712, in load
    return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
  File "/opt/conda/lib/python3.7/site-packages/torch/serialization.py", line 1046, in _load
    result = unpickler.load()
  File "/opt/conda/lib/python3.7/site-packages/torch/serialization.py", line 1039, in find_class
    return super().find_class(mod_name, name)
ModuleNotFoundError: No module named 'ultralytics.nn.modules.conv'; 'ultralytics.nn.modules' is not a package

检查 ultralytics包的版本，尽量是最新版本的，我因为原来是8.0.11，所以报错了。

查询问题以后发现，变成最新版本就没问题了

!pip install ultralytics==8.0.144

如果还是报错，要注意你运行的py文件要在yolo文件目录下面

由于yolov8训练出来的模型默认是保存在指定当前目录的runs/detect/train下面

如果要修改模型保存的路径可以在训练的时候加上project，如下所示：

yolo task=detect mode=train model=yolov8s.pt epochs=100 batch=2 data=datasets/helmet.yaml project=/project/train/models/resume

就可以实现训练的时候保存到指定的文件夹下面；

使用resume=True参数，就可以实现训练中断，然后再训练：

yolo task=detect mode=train model=E:/yolov8/ultralytics_ds_converter/ultralytics/please/train2/weights/best.pt epochs=100 batch=2 data=datasets/helmet.yaml project=E:/yolov8/ultralytics_ds_converter/ultralytics/please/  resume=True

但是要注意使用resume的模型要修改为你中断的模型里面，而且最好选择best模型，如果选择last.pt可能是损坏的，无法正常读取，会报下面的这个错误：

RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory

在使用C++里面的onnxruntime来运行yolov8s.onnxwen文件的时候，出现了报错信息：

Ort::Env(OrtLoggingLevel::ORT_LOGGING_LEVEL_ERROR, "Yolov5-Seg");
报错信息如下：
引发了异常: 读取访问权限冲突。
Ort::GetApi(...) 返回 nullptr。

我找到的解决方案如下：

这个一般会是dll冲突问题导致的，win系统特有的问题，原因在于win10的system32下面自带有一个onnxruntime的dll，优先级比环境变量添加的路径高导致的，你可以修改权限删除，或者你可以将onnx的相关dll拷贝到你项目的exe下面去运行看下。

如果出现了下面这个错误：

是路径的问题，把\变成/,检测onnx文件的路径。

我使用了opencv4.5.2和onnxruntime1.4.1

onnxruntime1.4.1

完整代码如下：

#include 
#include 
#include 

using namespace cv;
using namespace std;

std::string labels_txt_file = "classes.txt";
std::vector readClassNames();
std::vector readClassNames()
{
	std::vector classNames;

	std::ifstream fp(labels_txt_file);
	if (!fp.is_open())
	{
		printf("could not open file...\n");
		exit(-1);
	}
	std::string name;
	while (!fp.eof())
	{
		std::getline(fp, name);
		if (name.length())
			classNames.push_back(name);
	}
	fp.close();
	return classNames;
}

int main(int argc, char** argv) {
	std::vector labels = readClassNames();
	cv::Mat frame = cv::imread("E:/yolov8/dataset/images/.jpg");
	int ih = frame.rows;
	int iw = frame.cols;

	// 创建InferSession, 查询支持硬件设备
	// GPU Mode, 0 - gpu device id
	std::string onnxpath = "E:/yolov8/code/predict/predict/yolov8s.onnx";//E:/yolov8/ultralytics_ds_converter/ultralytics/runs/detect/train/weights/best.onnx
	std::wstring modelPath = std::wstring(onnxpath.begin(), onnxpath.end());
	Ort::SessionOptions session_options;
	Ort::Env env = Ort::Env(ORT_LOGGING_LEVEL_ERROR, "yolov8-onnx");

	session_options.SetGraphOptimizationLevel(ORT_ENABLE_BASIC);
	std::cout << "onnxruntime inference try to use GPU Device" << std::endl;
	OrtSessionOptionsAppendExecutionProvider_CUDA(session_options, 0);
	Ort::Session session_(env, modelPath.c_str(), session_options);

	std::vector input_node_names;
	std::vector output_node_names;

	size_t numInputNodes = session_.GetInputCount();
	size_t numOutputNodes = session_.GetOutputCount();
	Ort::AllocatorWithDefaultOptions allocator;
	input_node_names.reserve(numInputNodes);

	// 获取输入信息
	int input_w = 0;
	int input_h = 0;
	for (int i = 0; i < numInputNodes; i++) {
		auto input_name = session_.GetInputNameAllocated(i, allocator);
		input_node_names.push_back(input_name.get());
		Ort::TypeInfo input_type_info = session_.GetInputTypeInfo(i);
		auto input_tensor_info = input_type_info.GetTensorTypeAndShapeInfo();
		auto input_dims = input_tensor_info.GetShape();
		input_w = input_dims[3];
		input_h = input_dims[2];
		std::cout << "input format: w = " << input_w << "h:" << input_h << std::endl;
	}

	// 获取输出信息
	int output_h = 0;
	int output_w = 0;
	Ort::TypeInfo output_type_info = session_.GetOutputTypeInfo(0);
	auto output_tensor_info = output_type_info.GetTensorTypeAndShapeInfo();
	auto output_dims = output_tensor_info.GetShape();
	output_h = output_dims[1]; // 84
	output_w = output_dims[2]; // 8400
	std::cout << "output format : HxW = " << output_dims[1] << "x" << output_dims[2] << std::endl;
	for (int i = 0; i < numOutputNodes; i++) {
		auto out_name = session_.GetOutputNameAllocated(i, allocator);
		output_node_names.push_back(out_name.get());
	}
	std::cout << "input: " << input_node_names[0] << " output: " << output_node_names[0] << std::endl;

	// format frame
	int64 start = cv::getTickCount();
	int w = frame.cols;
	int h = frame.rows;
	int _max = std::max(h, w);
	cv::Mat image = cv::Mat::zeros(cv::Size(_max, _max), CV_8UC3);
	cv::Rect roi(0, 0, w, h);
	frame.copyTo(image(roi));

	// fix bug, boxes consistence!
	float x_factor = image.cols / static_cast(input_w);
	float y_factor = image.rows / static_cast(input_h);

	cv::Mat blob = cv::dnn::blobFromImage(image, 1 / 255.0, cv::Size(input_w, input_h), cv::Scalar(0, 0, 0), true, false);
	size_t tpixels = input_h * input_w * 3;
	std::array input_shape_info{ 1, 3, input_h, input_w };

	// set input data and inference
	auto allocator_info = Ort::MemoryInfo::CreateCpu(OrtDeviceAllocator, OrtMemTypeCPU);
	Ort::Value input_tensor_ = Ort::Value::CreateTensor(allocator_info, blob.ptr(), tpixels, input_shape_info.data(), input_shape_info.size());
	const std::array inputNames = { input_node_names[0].c_str() };
	const std::array outNames = { output_node_names[0].c_str() };
	std::vector ort_outputs;
	try {
		ort_outputs = session_.Run(Ort::RunOptions{ nullptr }, inputNames.data(), &input_tensor_, 1, outNames.data(), outNames.size());
	}
	catch (std::exception e) {
		std::cout << e.what() << std::endl;
	}

	// output data
	const float* pdata = ort_outputs[0].GetTensorMutableData();
	cv::Mat dout(output_h, output_w, CV_32F, (float*)pdata);
	cv::Mat det_output = dout.t(); // 8400x84

	// post-process
	std::vector boxes;
	std::vector classIds;
	std::vector confidences;

	for (int i = 0; i < det_output.rows; i++) {
		cv::Mat classes_scores = det_output.row(i).colRange(4, 84);
		cv::Point classIdPoint;
		double score;
		minMaxLoc(classes_scores, 0, &score, 0, &classIdPoint);

		// 置信度 0～1之间
		if (score > 0.25)
		{
			float cx = det_output.at(i, 0);
			float cy = det_output.at(i, 1);
			float ow = det_output.at(i, 2);
			float oh = det_output.at(i, 3);
			int x = static_cast((cx - 0.5 * ow) * x_factor);
			int y = static_cast((cy - 0.5 * oh) * y_factor);
			int width = static_cast(ow * x_factor);
			int height = static_cast(oh * y_factor);
			cv::Rect box;
			box.x = x;
			box.y = y;
			box.width = width;
			box.height = height;

			boxes.push_back(box);
			classIds.push_back(classIdPoint.x);
			confidences.push_back(score);
		}
	}

	// NMS
	std::vector indexes;
	cv::dnn::NMSBoxes(boxes, confidences, 0.25, 0.45, indexes);
	for (size_t i = 0; i < indexes.size(); i++) {
		int index = indexes[i];
		int idx = classIds[index];
		cv::rectangle(frame, boxes[index], cv::Scalar(0, 0, 255), 2, 8);
		cv::rectangle(frame, cv::Point(boxes[index].tl().x, boxes[index].tl().y - 20),
			cv::Point(boxes[index].br().x, boxes[index].tl().y), cv::Scalar(0, 255, 255), -1);
		putText(frame, labels[idx], cv::Point(boxes[index].tl().x, boxes[index].tl().y), cv::FONT_HERSHEY_PLAIN, 2.0, cv::Scalar(255, 0, 0), 2, 8);
		cv::imshow("YOLOv8+ONNXRUNTIME 对象检测演示", frame);
	}

	// 计算FPS render it
	float t = (cv::getTickCount() - start) / static_cast(cv::getTickFrequency());
	putText(frame, cv::format("FPS: %.2f", 1.0 / t), cv::Point(20, 40), cv::FONT_HERSHEY_PLAIN, 2.0, cv::Scalar(255, 0, 0), 2, 8);
	cv::imshow("YOLOv8+ONNXRUNTIME 对象检测演示", frame);
	cv::waitKey(0);

	session_options.release();
	session_.release();
	return 0;
}