【小白向】基于YOLOV5的数据集标注，训练，Windows/Linux/Jetson Nano多平台部署全流程¶

# anchors
anchors:
  - [10,13, 16,30, 33,23]  # P3/8
  - [30,61, 62,45, 59,119]  # P4/16
  - [116,90, 156,198, 373,326]  # P5/32

parser.add_argument('--noautoanchor', action='store_true', help='disable autoanchor check')

def check_anchors(dataset, model, thr=4.0, imgsz=640):
    # Check anchor fit to data, recompute if necessary
    print('\nAnalyzing anchors... ', end='')
    m = model.module.model[-1] if hasattr(model, 'module') else model.model[-1]  # Detect()
    shapes = imgsz * dataset.shapes / dataset.shapes.max(1, keepdims=True)
    scale = np.random.uniform(0.9, 1.1, size=(shapes.shape[0], 1))  # augment scale
    wh = torch.tensor(np.concatenate([l[:, 3:5] * s for s, l in zip(shapes * scale, dataset.labels)])).float()  # wh 

    def metric(k):  # compute metric
        r = wh[:, None] / k[None]
        x = torch.min(r, 1. / r).min(2)[0]  # ratio metric
        best = x.max(1)[0]  # best_x
        aat = (x > 1. / thr).float().sum(1).mean()  # anchors above threshold
        bpr = (best > 1. / thr).float().mean()  # best possible recall
        return bpr, aat

    bpr, aat = metric(m.anchor_grid.clone().cpu().view(-1, 2))
    print('anchors/target = %.2f, Best Possible Recall (BPR) = %.4f' % (aat, bpr), end='')
    if bpr < 0.98:  # threshold to recompute
        print('. Attempting to generate improved anchors, please wait...' % bpr)
        na = m.anchor_grid.numel() // 2  # number of anchors
        new_anchors = kmean_anchors(dataset, n=na, img_size=imgsz, thr=thr, gen=1000, verbose=False)
        new_bpr = metric(new_anchors.reshape(-1, 2))[0]
        if new_bpr > bpr:  # replace anchors
            new_anchors = torch.tensor(new_anchors, device=m.anchors.device).type_as(m.anchors)
            m.anchor_grid[:] = new_anchors.clone().view_as(m.anchor_grid)  # for inference
            m.anchors[:] = new_anchors.clone().view_as(m.anchors) / m.stride.to(m.anchors.device).view(-1, 1, 1)  # loss
            check_anchor_order(m)
            print('New anchors saved to model. Update model *.yaml to use these anchors in the future.')
        else:
            print('Original anchors better than new anchors. Proceeding with original anchors.')
    print('')  # newline

def kmean_anchors(path='./data/coco128.yaml', n=9, img_size=640, thr=4.0, gen=1000, verbose=True):
    """ Creates kmeans-evolved anchors from training dataset

        Arguments:
            path: path to dataset *.yaml, or a loaded dataset
            n: number of anchors
            img_size: image size used for training
            thr: anchor-label wh ratio threshold hyperparameter hyp['anchor_t'] used for training, default=4.0
            gen: generations to evolve anchors using genetic algorithm

        Return:
            k: kmeans evolved anchors

        Usage:
            from utils.general import *; _ = kmean_anchors()
    """

    thr = 1. / thr

    def metric(k, wh):  # compute metrics
        r = wh[:, None] / k[None]
        x = torch.min(r, 1. / r).min(2)[0]  # ratio metric
        # x = wh_iou(wh, torch.tensor(k))  # iou metric
        return x, x.max(1)[0]  # x, best_x

    def fitness(k):  # mutation fitness
        _, best = metric(torch.tensor(k, dtype=torch.float32), wh)
        return (best * (best > thr).float()).mean()  # fitness

    def print_results(k):
        k = k[np.argsort(k.prod(1))]  # sort small to large
        x, best = metric(k, wh0)
        bpr, aat = (best > thr).float().mean(), (x > thr).float().mean() * n  # best possible recall, anch > thr
        print('thr=%.2f: %.4f best possible recall, %.2f anchors past thr' % (thr, bpr, aat))
        print('n=%g, img_size=%s, metric_all=%.3f/%.3f-mean/best, past_thr=%.3f-mean: ' %
              (n, img_size, x.mean(), best.mean(), x[x > thr].mean()), end='')
        for i, x in enumerate(k):
            print('%i,%i' % (round(x[0]), round(x[1])), end=',  ' if i < len(k) - 1 else '\n')  # use in *.cfg
        return k

    if isinstance(path, str):  # *.yaml file
        with open(path) as f:
            data_dict = yaml.load(f, Loader=yaml.FullLoader)  # model dict
        from utils.datasets import LoadImagesAndLabels
        dataset = LoadImagesAndLabels(data_dict['train'], augment=True, rect=True)
    else:
        dataset = path  # dataset

    # Get label wh
    shapes = img_size * dataset.shapes / dataset.shapes.max(1, keepdims=True)
    wh0 = np.concatenate([l[:, 3:5] * s for s, l in zip(shapes, dataset.labels)])  # wh

    # Filter
    i = (wh0 < 3.0).any(1).sum()
    if i:
        print('WARNING: Extremely small objects found. '
              '%g of %g labels are < 3 pixels in width or height.' % (i, len(wh0)))
    wh = wh0[(wh0 >= 2.0).any(1)]  # filter > 2 pixels

    # Kmeans calculation
    print('Running kmeans for %g anchors on %g points...' % (n, len(wh)))
    s = wh.std(0)  # sigmas for whitening
    k, dist = kmeans(wh / s, n, iter=30)  # points, mean distance
    k *= s
    wh = torch.tensor(wh, dtype=torch.float32)  # filtered
    wh0 = torch.tensor(wh0, dtype=torch.float32)  # unflitered
    k = print_results(k)

    # Plot
    # k, d = [None] * 20, [None] * 20
    # for i in tqdm(range(1, 21)):
    #     k[i-1], d[i-1] = kmeans(wh / s, i)  # points, mean distance
    # fig, ax = plt.subplots(1, 2, figsize=(14, 7))
    # ax = ax.ravel()
    # ax[0].plot(np.arange(1, 21), np.array(d) ** 2, marker='.')
    # fig, ax = plt.subplots(1, 2, figsize=(14, 7))  # plot wh
    # ax[0].hist(wh[wh[:, 0]<100, 0],400)
    # ax[1].hist(wh[wh[:, 1]<100, 1],400)
    # fig.tight_layout()
    # fig.savefig('wh.png', dpi=200)

    # Evolve
    npr = np.random
    f, sh, mp, s = fitness(k), k.shape, 0.9, 0.1  # fitness, generations, mutation prob, sigma
    pbar = tqdm(range(gen), desc='Evolving anchors with Genetic Algorithm')  # progress bar
    for _ in pbar:
        v = np.ones(sh)
        while (v == 1).all():  # mutate until a change occurs (prevent duplicates)
            v = ((npr.random(sh) < mp) * npr.random() * npr.randn(*sh) * s + 1).clip(0.3, 3.0)
        kg = (k.copy() * v).clip(min=2.0)
        fg = fitness(kg)
        if fg > f:
            f, k = fg, kg.copy()
            pbar.desc = 'Evolving anchors with Genetic Algorithm: fitness = %.4f' % f
            if verbose:
                print_results(k)

    return print_results(k) 

python train_people.py

 python yolov5ToMsnhnet.py 

from PytorchToMsnhnet import *
Msnhnet.Export = True
from models.experimental import attempt_load
import torch

weights     = "weights/best.pt" # pt文件
msnhnetPath = "yolov5m.msnhnet" # 导出.msnhnet文件
msnhbinPath = "yolov5m.msnhbin" # 导出.msnhbin文件

model = attempt_load(weights, "cpu") 
model.eval() # cpu模式，推理模式

img = torch.rand(512*512*3).reshape(1,3,512,512) #生成随机推理数据

tans(model,img,msnhnetPath,msnhbinPath) #模型转换

  - CMAKE_INSTALL_PREFIX #指定安装位置,如: D:/libs/opencv
  - CPU_BASELINE #选择AVX2(如果CPU支持AVX2加速)
  - BUILD_TESTS #取消勾选

 - CMAKE_INSTALL_PREFIX #指定安装位置,如: D:/libs/Msnhnet
 - BUILD_EXAMPLE      #构建示例
 - BUILD_SHARED_LIBS  #构建动态链接库
 - BUILD_USE_CUDNN    #使用CUDNN
 - BUILD_USE_GPU      #使用GPU
 - BUILD_USE_OPENCV   #使用OPENCV
 - ENABLE_OMP         #使用OMP
 - OMP_MAX_THREAD     #使用最大核心数

yolov3_gpu D:/models
yolov3_gpu_fp16 D:/models #fp16推理

git clone https://github.com/msnh2012/MsnhnetSharp

cmake_minimum_required(VERSION 3.15)
project(yolov5m_msnhnet
        LANGUAGES CXX C CUDA
        VERSION 1.0)

find_package(OpenCV REQUIRED)
find_package(Msnhnet REQUIRED)
find_package(OpenMP REQUIRED)

add_executable(yolov5m_msnhnet yolov5m_msnhnet.cpp)

target_include_directories(yolov5m_msnhnet PUBLIC ${Msnhnet_INCLUDE_DIR})
target_link_libraries(yolov5m_msnhnet PUBLIC ${OpenCV_LIBS} Msnhnet)

#include <iostream>
#include "Msnhnet/net/MsnhNetBuilder.h"
#include "Msnhnet/io/MsnhIO.h"
#include "Msnhnet/config/MsnhnetCfg.h"
#include "Msnhnet/utils/MsnhOpencvUtil.h"

void yolov5sGPUOpencv(const std::string& msnhnetPath, const std::string& msnhbinPath, const std::string& imgPath,  const std::string& labelsPath)
{
    try
    {
        Msnhnet::NetBuilder  msnhNet;
        Msnhnet::NetBuilder::setOnlyGpu(true);
        //msnhNet.setUseFp16(true); //开启使用FP16推理
        msnhNet.buildNetFromMsnhNet(msnhnetPath);
        std::cout<<msnhNet.getLayerDetail();
        msnhNet.loadWeightsFromMsnhBin(msnhbinPath);
        std::vector<std::string> labels ;
        Msnhnet::IO::readVectorStr(labels, labelsPath.data(), "\n");
        Msnhnet::Point2I inSize = msnhNet.getInputSize();

        std::vector<float> img;
        std::vector<std::vector<Msnhnet::YoloBox>> result;

        img = Msnhnet::OpencvUtil::getPaddingZeroF32C3(imgPath, cv::Size(inSize.x,inSize.y));
        for (size_t i = 0; i < 10; i++)
        {
            auto st = Msnhnet::TimeUtil::startRecord();
            result = msnhNet.runYoloGPU(img);
            std::cout<<"time  : " << Msnhnet::TimeUtil::getElapsedTime(st) <<"ms"<<std::endl<<std::flush;
        }

        cv::Mat org = cv::imread(imgPath);
        Msnhnet::OpencvUtil::drawYoloBox(org,labels,result,inSize);
        cv::imshow("test",org);
        cv::waitKey();
    }
    catch (Msnhnet::Exception ex)
    {
        std::cout<<ex.what()<<" "<<ex.getErrFile() << " " <<ex.getErrLine()<< " "<<ex.getErrFun()<<std::endl;
    }
}

int main(int argc, char** argv)
{
    std::string msnhnetPath = "yolov5m.msnhnet";
    std::string msnhbinPath = "yolov5m.msnhbin";
    std::string labelsPath  = "labels.txt";
    std::string imgPath = "bus.jpg";

    yolov5sGPUOpencv(msnhnetPath, msnhbinPath, imgPath,labelsPath);
    getchar();

    return 0;
}

sudo apt-get install build-essential

sudo apt-get install libopencv 

 - CMAKE_INSTALL_PREFIX #指定安装位置,如: D:/libs/Msnhnet
 - BUILD_EXAMPLE      #构建示例
 - BUILD_SHARED_LIBS  #构建动态链接库
 - BUILD_USE_CUDNN    #使用CUDNN
 - BUILD_USE_GPU      #使用GPU
 - BUILD_USE_NEON     #使用neon加速
 - BUILD_USE_OPENCV   #使用OPENCV
 - ENABLE_OMP         #使用OMP
 - OMP_MAX_THREAD     #使用最大核心数

make -j
sudo make install

sudo gedit /etc/ld.so.conf.d/usr.confg
# 添加: /usr/local/lib
sudo ldconfig

cd /usr/local/bin
yolov3_gpu /home/xxx/models
yolov3_gpu_fp16 /home/xxx/models #fp16推理

cmake_minimum_required(VERSION 3.15)
project(yolov5m_msnhnet
        LANGUAGES CXX C CUDA
        VERSION 1.0)

find_package(OpenCV REQUIRED)
find_package(Msnhnet REQUIRED)
find_package(OpenMP REQUIRED)

add_executable(yolov5m_msnhnet yolov5m_msnhnet.cpp)

target_include_directories(yolov5m_msnhnet PUBLIC ${Msnhnet_INCLUDE_DIR})
target_link_libraries(yolov5m_msnhnet PUBLIC ${OpenCV_LIBS} Msnhnet)

#include <iostream>
#include "Msnhnet/net/MsnhNetBuilder.h"
#include "Msnhnet/io/MsnhIO.h"
#include "Msnhnet/config/MsnhnetCfg.h"
#include "Msnhnet/utils/MsnhOpencvUtil.h"


void yolov5sGPUOpencv(const std::string& msnhnetPath, const std::string& msnhbinPath, const std::string& imgPath,  const std::string& labelsPath)
{
    try
    {
        Msnhnet::NetBuilder  msnhNet;
        Msnhnet::NetBuilder::setOnlyGpu(true);
        //msnhNet.setUseFp16(true); //开启使用FP16推理
        msnhNet.buildNetFromMsnhNet(msnhnetPath);
        std::cout<<msnhNet.getLayerDetail();
        msnhNet.loadWeightsFromMsnhBin(msnhbinPath);
        std::vector<std::string> labels ;
        Msnhnet::IO::readVectorStr(labels, labelsPath.data(), "\n");
        Msnhnet::Point2I inSize = msnhNet.getInputSize();

        std::vector<float> img;
        std::vector<std::vector<Msnhnet::YoloBox>> result;

        img = Msnhnet::OpencvUtil::getPaddingZeroF32C3(imgPath, cv::Size(inSize.x,inSize.y));
        for (size_t i = 0; i < 10; i++)
        {
            auto st = Msnhnet::TimeUtil::startRecord();
            result = msnhNet.runYoloGPU(img);
            std::cout<<"time  : " << Msnhnet::TimeUtil::getElapsedTime(st) <<"ms"<<std::endl<<std::flush;
        }

        cv::Mat org = cv::imread(imgPath);
        Msnhnet::OpencvUtil::drawYoloBox(org,labels,result,inSize);
        cv::imshow("test",org);
        cv::waitKey();
    }
    catch (Msnhnet::Exception ex)
    {
        std::cout<<ex.what()<<" "<<ex.getErrFile() << " " <<ex.getErrLine()<< " "<<ex.getErrFun()<<std::endl;
    }
}

int main(int argc, char** argv)
{
    std::string msnhnetPath = "../yolov5m.msnhnet";
    std::string msnhbinPath = "../yolov5m.msnhbin";
    std::string labelsPath  = "../labels.txt";
    std::string imgPath = "../bus.jpg";

    yolov5sGPUOpencv(msnhnetPath, msnhbinPath, imgPath,labelsPath);
    getchar();

    return 0;
}

mkdir build
cd build 
make 
./yolov5m_msnhnet