1. 創(chuàng)建虛擬環(huán)境

```python
conda create -n yolos python=3.11
# 激活yolos 環(huán)境，后續(xù)的安裝都在里面進行
conda activate yolos

2. 通過git 安裝 ultralytics

# 沒有g(shù)it的話要安裝git
conda install git

# D:  進入D盤
D:
mkdir yolos
cd yolos

# Clone the ultralytics repository
git clone https://github.com/ultralytics/ultralytics

# Navigate to the cloned directory
cd ultralytics

# Install the package in editable mode for development
pip install -e .    //最后的“.”不可省略

# 通過該命令安裝的torch 是cpu版本，如果需要安裝gpu，需要先卸載掉，然后安裝
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117  //這一步驟比較慢

# torch 安裝完成后，可以執(zhí)行如下命令，進行快速安裝
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt

# 后面有些代碼需要pytest，也要安裝一下
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple pytest

# 如果需要安裝onnx，也可以安裝一下
conda install onnx

3. 安裝完成之后，通過以下代碼測試下環(huán)境配置是否正確

# DemoTest.py
# yolo predict model=yolov8n-seg.pt source='https://youtu.be/Zgi9g1ksQHc' imgsz=320   通過該命令可自動下載不同模型
from multiprocessing import freeze_support
from ultralytics import YOLO

def main():
    # Load a model    
    model = YOLO("yolov8n.yaml")  # build a new model from scratch    
    model = YOLO("yolov8n.pt")  # load a pretrained model (recommended for training)   
     # Use the model    
     model.train(data="coco128.yaml", epochs=3)  # train the model   
      metrics = model.val()  # evaluate model performance on the validation set    
      results = model("https://ultralytics.com/images/bus.jpg")  # predict on an image    
      # path = model.export(format="onnx")  # export the model to ONNX format
      
if __name__ == '__main__':
    freeze_support()
    main()

運行過程中，會提示下載coco128.zip,和yolov8n.pt，模型與py文件放在同一目錄下，如果網(wǎng)速快的話，自己下載就好，下載慢的話，下面是網(wǎng)盤地址：

yolov8 檢測預訓練模型百度網(wǎng)盤：鏈接：https://pan.baidu.com/s/1L5q1sdtBmq0FcuX6t1SvIg  提取碼：ix9e
coco128.zip  百度網(wǎng)盤： 鏈接：https://pan.baidu.com/s/1UMdrWcY49jfCVvGvTMm8xg 提取碼：rqd0

測試結(jié)果路徑：ultralytics\runs\detect\val，里面存儲了運行的結(jié)果，這樣環(huán)境就算是配置好了。

結(jié)果如下：

4. 安裝labelImg標注軟件

# 下載源代碼
git clone https://github.com/HumanSignal/labelImg.git
# 創(chuàng)建labelImg虛擬環(huán)境，lebelImg 需要低版本的python,我這里安裝3.7
conda create -n labelImg37 python=3.7
# 激活環(huán)境
conda activate labelImg37
# 安裝依賴庫
conda install pyqt=5
conda install -c anaconda lxml
# 將qrc轉(zhuǎn)換成可調(diào)用的py
pyrcc5 -o libs/resources.py resources.qrc  
# 直接運行會報錯 'pyrcc5' 不是內(nèi)部或外部命令，也不是可運行的程序；因為從anaconda 中安裝的pyqt不包含pyrcc5
# 需要從cmd直接安裝
pip install pyqt5_tools -i https://pypi.tuna.tsinghua.edu.cn/simple
# 然后再執(zhí)行下一句
pyrcc5 -o libs/resources.py resources.qrc  
#然后執(zhí)行下一句彈出窗口
python labelImg.py
# python labelImg.py [IMAGE_PATH] [PRE-DEFINED CLASS FILE]

# 也可以直接通過pip安裝
pip3 install labelImg
# 啟動
labelImg

5. 使用labelImg進行標注，圖片使用上面的coco128

首先創(chuàng)建一個文件夾：cocoImages, 里面分別創(chuàng)建2個文件夾，images用來放置標注圖片, vocLabels 用來放置標注文件

5.1 點擊“打開目錄”選擇存儲圖像的文件夾進行標注，右下角會出現(xiàn)圖像列表

5.2 選擇“創(chuàng)建區(qū)塊”，在圖像上對目標進行標注，然后填入類別，每張圖片皆可標記多個目標

5.3 每一張圖片標注完后，軟件會提示進行保存，點擊Yes即可；

5.4 標記完后的文件如圖所示；

5.5 將xml文件放入vocLabels文件夾中；

6. 將數(shù)據(jù)轉(zhuǎn)換成yolo需要的格式

首先將11行中的classes改為自己標注的類別，然后執(zhí)行下代碼生成相應的文件夾，接著將圖像copy到JPEGImages下，labels copy到Annotations下面，再次執(zhí)行一次該代碼即可。

import xml.etree.ElementTree as ET
import pickle
import os
from os import listdir, getcwd
from os.path import join
import random

# classes=["aeroplane", 'bicycle', 'bird', 'boat', 'bottle', 'bus', 'car', 'cat', 'chair', 'cow', 'diningtable', 'dog',
#            'horse', 'motorbike', 'person', 'pottedplant', 'sheep', 'sofa', 'train', 'tvmonitor']  # class names

classes = ["person", 'cup', 'umbrella']


def clear_hidden_files(path):
    dir_list = os.listdir(path)
    for i in dir_list:
        abspath = os.path.join(os.path.abspath(path), i)
        if os.path.isfile(abspath):
            if i.startswith("._"):
                os.remove(abspath)
        else:
            clear_hidden_files(abspath)


def convert(size, box):
    dw = 1. / size[0]
    dh = 1. / size[1]
    x = (box[0] + box[1]) / 2.0
    y = (box[2] + box[3]) / 2.0
    w = box[1] - box[0]
    h = box[3] - box[2]
    x = x * dw
    w = w * dw
    y = y * dh
    h = h * dh
    return x, y, w, h


def convert_annotation(image_id, voc_labels, yolo_labels):
    in_file = open(os.path.join(voc_labels + '%s.xml') % image_id)
    out_file = open(os.path.join(yolo_labels + '%s.txt') % image_id, 'w')
    tree = ET.parse(in_file)
    root = tree.getroot()
    size = root.find('size')
    w = int(size.find('width').text)
    h = int(size.find('height').text)

    for obj in root.iter('object'):
        difficult = obj.find('difficult').text
        cls = obj.find('name').text
        if cls not in classes or int(difficult) == 1:
            continue
        cls_id = classes.index(cls)
        xml_box = obj.find('bndbox')
        b = (float(xml_box.find('xmin').text), float(xml_box.find('xmax').text), float(xml_box.find('ymin').text),
             float(xml_box.find('ymax').text))
        bb = convert((w, h), b)
        out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
    in_file.close()
    out_file.close()


if __name__ == '__main__':
    # 獲取當前路徑
    wd = os.getcwd()
    # 創(chuàng)建相應VOC模式文件夾
    voc_path = os.path.join(wd, "voc_dataset")
    if not os.path.isdir(voc_path):
        os.mkdir(voc_path)

    annotation_dir = os.path.join(voc_path, "Annotations/")
    if not os.path.isdir(annotation_dir):
        os.mkdir(annotation_dir)
    clear_hidden_files(annotation_dir)

    image_dir = os.path.join(voc_path, "JPEGImages/")
    if not os.path.isdir(image_dir):
        os.mkdir(image_dir)
    clear_hidden_files(image_dir)

    voc_file_dir = os.path.join(voc_path, "ImageSets/")
    if not os.path.isdir(voc_file_dir):
        os.mkdir(voc_file_dir)

    voc_file_dir = os.path.join(voc_file_dir, "Main/")
    if not os.path.isdir(voc_file_dir):
        os.mkdir(voc_file_dir)

    VOC_train_file = open(os.path.join(voc_path, "ImageSets/Main/train.txt"), 'w')
    VOC_test_file = open(os.path.join(voc_path, "ImageSets/Main/test.txt"), 'w')
    VOC_train_file.close()
    VOC_test_file.close()

    if not os.path.exists(os.path.join(voc_path, 'Labels/')):
        os.makedirs(os.path.join(voc_path, 'Labels'))

    train_file = open(os.path.join(voc_path, "2007_train.txt"), 'a')
    test_file = open(os.path.join(voc_path, "2007_test.txt"), 'a')
    VOC_train_file = open(os.path.join(voc_path, "ImageSets/Main/train.txt"), 'a')
    VOC_test_file = open(os.path.join(voc_path, "ImageSets/Main/test.txt"), 'a')

    image_list = os.listdir(image_dir)  # list image files
    probo = random.randint(1, 100)
    print("Probobility: %d" % probo)
    for i in range(0, len(image_list)):
        path = os.path.join(image_dir, image_list[i])
        if os.path.isfile(path):
            image_path = image_dir + image_list[i]
            image_name = image_list[i]
            (name_without_extent, extent) = os.path.splitext(os.path.basename(image_path))
            voc_name_without_extent, voc_extent = os.path.splitext(os.path.basename(image_name))
            annotation_name = name_without_extent + '.xml'
            annotation_path = os.path.join(annotation_dir, annotation_name)
        probo = random.randint(1, 100)
        print("Probobility: %d" % probo)
        if (probo < 75):
            if os.path.exists(annotation_path):
                train_file.write(image_path + '\n')
                VOC_train_file.write(voc_name_without_extent + '\n')
                yolo_labels_dir = os.path.join(voc_path, 'Labels/')
                convert_annotation(name_without_extent, annotation_dir, yolo_labels_dir)
        else:
            if os.path.exists(annotation_path):
                test_file.write(image_path + '\n')
                VOC_test_file.write(voc_name_without_extent + '\n')
                yolo_labels_dir =os.path.join(voc_path, 'Labels/')
                convert_annotation(name_without_extent, annotation_dir, yolo_labels_dir)

    train_file.close()
    test_file.close()
    VOC_train_file.close()
    VOC_test_file.close()

7. 對數(shù)據(jù)集進行劃分

import os
import shutil
import random
ratio=0.1
img_dir='./voc_dataset/JPEGImages' #圖片路徑
label_dir='./voc_dataset/Labels'#生成的yolo格式的數(shù)據(jù)存放路徑
train_img_dir='./voc_dataset/images/train2017'#訓練集圖片的存放路徑
val_img_dir='./voc_dataset/images/val2017'
train_label_dir='./voc_dataset/labels/train2017'#訓練集yolo格式數(shù)據(jù)的存放路徑
val_label_dir='./voc_dataset/labels/val2017'
if not os.path.exists(train_img_dir):
    os.makedirs(train_img_dir)
if not os.path.exists(val_img_dir):
    os.makedirs(val_img_dir)
if not os.path.exists(train_label_dir):
    os.makedirs(train_label_dir)
if not os.path.exists(val_label_dir):
    os.makedirs(val_label_dir)
names=os.listdir(img_dir)
val_names=random.sample(names,int(len(names)*ratio))

cnt_1=0
cnt_2=0
for name in names:
    if name in val_names:
        #cnt_1+=1
        #if cnt_1>100:
            #break
        shutil.copy(os.path.join(img_dir,name),os.path.join(val_img_dir,name))
        shutil.copy(os.path.join(label_dir, name[:-4]+'.txt'), os.path.join(val_label_dir, name[:-4]+'.txt'))
    else:
        #cnt_2+=1
        #if cnt_2>1000:
            #break
        shutil.copy(os.path.join(img_dir, name), os.path.join(train_img_dir, name))
        shutil.copy(os.path.join(label_dir, name[:-4] + '.txt'), os.path.join(train_label_dir, name[:-4] + '.txt'))

執(zhí)行完第七個步驟后，數(shù)據(jù)集的文件分布如下所示，其中，images,Labels中的文件即yolov8訓練時所需要的：

8.訓練

8.1 如果運行的時候出現(xiàn)如下報錯，進入虛擬環(huán)境中搜索libiomp5md.dll，刪掉一個即可

OMP: Error #15: Initializing libiomp5md.dll, but found libiomp5md.dll already initialized

8.2 訓練時需要修改的文件如下，修改文件的路徑如下：

D:\yolos\ultralytics\ultralytics\cfg\datasets\myVOC.yaml

# Ultralytics YOLO ??, AGPL-3.0 license# PASCAL VOC dataset http://host.robots.ox.ac.uk/pascal/VOC by University of Oxford# Example usage: yolo train data=VOC.yaml# parent# ├── ultralytics# └── datasets#     └── VOC  ← downloads here (2.8 GB)# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]path: ../voc_dataset/
train: # train images (relative to 'path')  16551 images  - images/train2017
val: # val images (relative to 'path')  4952 images  - images/train2017
test: # test images (optional)  - images/train2017

# Classesnames:
  0: person
  1: cup
  2: umbrella

網(wǎng)絡(luò)配置參數(shù)：

?D:\yolos\ultralytics\ultralytics\cfg\models\v8\yolov8.yaml

下面的文件是網(wǎng)絡(luò)訓練時的參數(shù)，可以進行修改，如果訓練的次數(shù)少，沒有結(jié)果，可以修改該配置里面的conf

D:\yolos\ultralytics\ultralytics\cfg\default.yaml

8.3 訓練

修改完成后，訓練完整代碼如下：

from multiprocessing import freeze_support

import cv2

from ultralytics import YOLO

def main():
    # Load a model
    model = YOLO("yolov8n.yaml")  # build a new model from scratch
    model = YOLO("yolov8n.pt")  # load a pretrained model (recommended for training)

    # Use the model
    model.train(data="myVOC.yaml", epochs=100)  # train the model
    metrics = model.val()  # evaluate model performance on the validation set

if __name__ == '__main__':
    freeze_support()
    main()

訓練結(jié)果,
訓練的結(jié)果和模型，在文件夾runs中：

9. Predict 預測

預測完整代碼如下：
代碼如下，運行的時候需要注意修改模型的路徑

from multiprocessing import freeze_support

import cv2

from ultralytics import YOLO

def main():
    # Load a model
    model = YOLO("yolov8n.yaml")  # build a new model from scratch
    model = YOLO("runs/detect/train18/weights/best.pt")  # load a pretrained model (recommended for training)
    results = model("000000000036.jpg")  # predict on an image
    path = model.export(format="onnx")  # export the model to ONNX format
    # Process results list

    for res in results:
        boxes = res.boxes  # Boxes object for bbox outputs
        masks = res.masks  # Masks object for segmentation masks outputs
        keypoints = res.keypoints  # Keypoints object for pose outputs
        probs = res.probs  # Probs object for classification outputs
        res_plotted = res.plot()
        cv2.namedWindow("yolov8_result", cv2.WINDOW_NORMAL)
        cv2.imshow("yolov8_result", res_plotted)
        cv2.waitKey(0)

if __name__ == '__main__':
    freeze_support()
    main()

最終輸出結(jié)果如下：

10. ONNX

path = model.export(format="onnx")  # export the model to ONNX format    

這句代碼輸出onnx格式的模型，可以通過提示查看網(wǎng)絡(luò)結(jié)果：
運行完成后，terminal中會出現(xiàn)以下提示，可以點擊網(wǎng)址，然后從網(wǎng)址中打開路徑中的best.onnx,即可查看網(wǎng)絡(luò)模型。

11. 在此基礎(chǔ)上運行yolov5

Yolov5 也可以直接使用，注意ultralytics路徑，可以單獨拷貝一份出來，放在正確的路徑中,下面時我這邊的文件分布；
若已經(jīng)安裝過git，仍然報錯，那就需要去安裝目錄中找到git.exe, 然后手動添加到環(huán)境變量。?C:\Users\86942.conda\envs\yolos\Library\mingw64\bin\git.exe

文章來源地址http://www.zghlxwxcb.cn/news/detail-646782.html

到了這里，關(guān)于基于Yolov8與LabelImg訓練自己數(shù)據(jù)的完整流程的文章就介紹完了。如果您還想了解更多內(nèi)容，請在右上角搜索TOY模板網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關(guān)文章，希望大家以后多多支持TOY模板網(wǎng)！