一、环境配置
1.1 创建python环境
1.conda create -n yolov8 python=3.8 //创建环境
2.conda env list //查看所有环境及其路径1.2 下载yolov8源码
1.下载路径，点击zip下载即可，也可直接git克隆。
https://github.com/ultralytics/ultralytics/2.下载zip后解压，使用pycharm打开该文件，如图所示
3.进入python解释器界面，选择环境
4.。。。pycharm2024识别不了conda环境下的python.exe。换pycharm2022
换pycharm版本后，添加环境
如下所示即为成功：
二、下载Pytorch
折腾几天，各种换源都下载的cpu版本。
最终在阿里云下找到，经过折腾最终将CUDA版本更换为11.8
https://mirrors.aliyun.com/pytorch-wheels/cu118/在链接中下载t对应版本的torch，torchvision，torchaudio的whl文件到目录下。
最后使用需要使用的conda环境cd到该目录下
pip install "torch-xxxxx.whl"
pip install "torchvision-xxxxx.whl"
pip install "torchaudio-xxxxx.whl"三、开始训练
需要再安装一个包，使用以下命令
pip install ultralytics准备coco128数据集，以及yolov8n.pt或yolov8s.pt,如下图所示：新建目录dataset，修改coco128.yaml的path为准备好的数据集路径。权重放在新建的weight目录下。
随后使用以下命令，其中data代表数据集，model代表使用的权重，epoch代表迭代多少次，batch为一轮训练中每一次放入图片数量，越大越快效果越好，但是对性能要求越高，device代表使用什么设备进行训练。
yolo train data=ultralytics/cfg/datasets/coco128.yaml model=weight/yolov8s.pt epochs=100 batch=4 device=0训练结束后可在以下路径中找到best.pt
四、使用自己的数据集进行训练
1.在coco128数据集相同的位置导入图像，并且进行过图像标注（我使用的是VOC格式），如下图中ZJHH目录下的Annotations和images，其中Annotations下为标注后的xml文件，images为原图像。
2.在根目录下新建split_train_val.py文件，为以下内容，运行文件后得到得到trainval.txt、val.txt、test.txt
import os
import random
trainval_percent = 0.1
train_percent = 0.9
xmlfilepath = 'dataset/DataSet/ZJHH/Annotations'
txtsavepath = 'dataset/DataSet/ZJHH/imgSets'
total_xml = os.listdir(xmlfilepath)
num = len(total_xml)
list = range(num)
tv = int(num * trainval_percent)
tr = int(tv * train_percent)
trainval = random.sample(list, tv)
train = random.sample(trainval, tr)
ftrainval = open('dataset/DataSet/ZJHH/imgSets/trainval.txt', 'w')
ftest = open('dataset/DataSet/ZJHH/imgSets/test.txt', 'w')
ftrain = open('dataset/DataSet/ZJHH/imgSets/train.txt', 'w')
fval = open('dataset/DataSet/ZJHH/imgSets/val.txt', 'w')
for i in list:
name = total_xml[i][:-4] + '\n'
if i in trainval:
ftrainval.write(name)
if i in train:
ftest.write(name)
else:
fval.write(name)
else:
ftrain.write(name)
ftrainval.close()
ftrain.close()
fval.close()
ftest.close()如下图所示，得到以下几个文件
3.在根目录下创建voc_label.py文件，为以下内容，运行后对于每一张图像，程序会生成一个相应的文本文件存放在labels目录下。
import xml.etree.ElementTree as ET
import os
from os import getcwd
sets = ['train', 'val', 'test']
classes = ['批号']
abs_path = os.getcwd()
print(abs_path)
def convert(size, box):
dw = 1. / (size[0])
dh = 1. / (size[1])
x = (box[0] + box[1]) / 2.0 - 1
y = (box[2] + box[3]) / 2.0 - 1
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):
in_file = open('dataset/DataSet/ZJHH/Annotations/%s.xml' % (image_id), encoding='UTF-8')
out_file = open('dataset/DataSet/ZJHH/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
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)
xmlbox = obj.find('bndbox')
b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text),
float(xmlbox.find('ymax').text))
b1, b2, b3, b4 = b
# 标注越界修正
if b2 > w:
b2 = w
if b4 > h:
b4 = h
b = (b1, b2, b3, b4)
bb = convert((w, h), b)
out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
wd = getcwd()
for image_set in sets:
if not os.path.exists('dataset/DataSet/ZJHH/labels'):
os.makedirs('dataset/DataSet/ZJHH/labels')
image_ids = open('dataset/DataSet/ZJHH/imgSets/%s.txt' % (image_set)).read().strip().split()
list_file = open('dataset/DataSet/ZJHH/%s.txt' % (image_set), 'w')
for image_id in image_ids:
list_file.write(abs_path + 'dataset/DataSet/ZJHH/images/%s.jpg\n' % (image_id))
convert_annotation(image_id)
list_file.close()运行后得到以下文件，其中labels下为VOC格式转化为YOLO格式的标注数据
4.在ultralytics/cfg/datasets下创建“zjhhBN.yaml”为以下内容
train: dataset/DataSet/ZJHH/train.txt
val: dataset/DataSet/ZJHH/val.txt
test: dataset/DataSet/ZJHH/test.txt
nc: 1 # 填写类别数
names: ["批号"] # 要和voc_label.py中的类别数目顺序一模一样
# train,val,test的路径根据自己情况而定5.也可使用coco128的方式进行数据集格式的转换，在Annotations同目录下新建YoLolabels文件夹以及在根目录下新建应该vocToYoLO.py文件，内容如下：
import xml.etree.ElementTree as ET
import pickle
import os
from os import listdir, getcwd
from os.path import join
import random
from shutil import copyfile
classes = ['批号'] # 划分的类别，依实际情况而定
# classes=["ball"]
TRAIN_RATIO = 80 # 训练集与数据集划分的比例
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): # 边界框的转换
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):
in_file = open('dataset/DataSet/ZJHH/Annotations/%s.xml' % image_id,encoding='utf-8') # xml路径
out_file = open('dataset/DataSet/ZJHH/YoLolabels/%s.txt' % image_id, 'w',encoding='utf-8')
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)
xmlbox = obj.find('bndbox')
b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text),
float(xmlbox.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()
wd = os.getcwd()
wd = os.getcwd()
data_base_dir = os.path.join(wd, "dataset/DataSet/") #数据集目录
if not os.path.isdir(data_base_dir):
os.mkdir(data_base_dir)
work_sapce_dir = os.path.join(data_base_dir, "ZJHH/") #xml文件及图片上级目录
if not os.path.isdir(work_sapce_dir):
os.mkdir(work_sapce_dir)
annotation_dir = os.path.join(work_sapce_dir, "Annotations/") #xml文件目录
if not os.path.isdir(annotation_dir):
os.mkdir(annotation_dir)
clear_hidden_files(annotation_dir)
image_dir = os.path.join(work_sapce_dir, "images/") #图片目录
if not os.path.isdir(image_dir):
os.mkdir(image_dir)
clear_hidden_files(image_dir)
yolo_labels_dir = os.path.join(work_sapce_dir, "YoLolabels/") #数据集格式转换存放目录
if not os.path.isdir(yolo_labels_dir):
os.mkdir(yolo_labels_dir)
clear_hidden_files(yolo_labels_dir)
yolov5_images_dir = os.path.join(data_base_dir, "images/")
if not os.path.isdir(yolov5_images_dir):
os.mkdir(yolov5_images_dir)
clear_hidden_files(yolov5_images_dir)
yolov5_labels_dir = os.path.join(data_base_dir, "labels/")
if not os.path.isdir(yolov5_labels_dir):
os.mkdir(yolov5_labels_dir)
clear_hidden_files(yolov5_labels_dir)
yolov5_images_train_dir = os.path.join(yolov5_images_dir, "train/")
if not os.path.isdir(yolov5_images_train_dir):
os.mkdir(yolov5_images_train_dir)
clear_hidden_files(yolov5_images_train_dir)
yolov5_images_test_dir = os.path.join(yolov5_images_dir, "val/")
if not os.path.isdir(yolov5_images_test_dir):
os.mkdir(yolov5_images_test_dir)
clear_hidden_files(yolov5_images_test_dir)
yolov5_labels_train_dir = os.path.join(yolov5_labels_dir, "train/")
if not os.path.isdir(yolov5_labels_train_dir):
os.mkdir(yolov5_labels_train_dir)
clear_hidden_files(yolov5_labels_train_dir)
yolov5_labels_test_dir = os.path.join(yolov5_labels_dir, "val/")
if not os.path.isdir(yolov5_labels_test_dir):
os.mkdir(yolov5_labels_test_dir)
clear_hidden_files(yolov5_labels_test_dir)
train_file = open(os.path.join(wd, "yolov5_train.txt"), 'w')
test_file = open(os.path.join(wd, "yolov5_val.txt"), 'w')
train_file.close()
test_file.close()
train_file = open(os.path.join(wd, "yolov5_train.txt"), 'a')
test_file = open(os.path.join(wd, "yolov5_val.txt"), 'a')
list_imgs = os.listdir(image_dir) # list image files
prob = random.randint(1, 100)
print("Probability: %d" % prob)
for i in range(0, len(list_imgs)):
path = os.path.join(image_dir, list_imgs[i])
if os.path.isfile(path):
image_path = image_dir + list_imgs[i]
voc_path = list_imgs[i]
(nameWithoutExtention, extention) = os.path.splitext(os.path.basename(image_path))
(voc_nameWithoutExtention, voc_extention) = os.path.splitext(os.path.basename(voc_path))
annotation_name = nameWithoutExtention + '.xml'
annotation_path = os.path.join(annotation_dir, annotation_name)
label_name = nameWithoutExtention + '.txt'
label_path = os.path.join(yolo_labels_dir, label_name)
prob = random.randint(1, 100)
print("Probability: %d" % prob)
if (prob < TRAIN_RATIO): # train dataset
if os.path.exists(annotation_path):
train_file.write(image_path + '\n')
convert_annotation(nameWithoutExtention) # convert label
copyfile(image_path, yolov5_images_train_dir + voc_path)
copyfile(label_path, yolov5_labels_train_dir + label_name)
else: # test dataset
if os.path.exists(annotation_path):
test_file.write(image_path + '\n')
convert_annotation(nameWithoutExtention) # convert label
copyfile(image_path, yolov5_images_test_dir + voc_path)
copyfile(label_path, yolov5_labels_test_dir + label_name)
train_file.close()
test_file.close()
注：若类别中含有中文，该段代码encoding应为utf-8.否则会出现格式转换后文件为空的情况。
in_file = open('dataset/DataSet/ZJHH/Annotations/%s.xml' % image_id,encoding='utf-8') # xml路径
out_file = open('dataset/DataSet/ZJHH/YoLolabels/%s.txt' % image_id, 'w',encoding='utf-8')6.运行该代码后，会生成以下文件及文件夹
其中DataSet/images/train、val下存放按数据集验证集随机分类的图片，labels下存放转换后的数据集标注，内容如下：，若为空则说明转换失败，可检查中文编码问题。
随后使用以下命令进行训练
yolo train data=ultralytics/cfg/datasets/zjhhBN.yaml model=weight/yolov8s.pt epochs=100 batch=4 device=0因为类型包含中文，所以会自动下载应该ttf文件，因为下载太慢，也可以复制网址下载后复制到字符路径下，在下图红圈位置。
开始训练
训练完成后可在run目录下找到训练的模型
扫一扫
1192
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
