yolo的笔记

检查是否安装好环境

import torch
import torchvision

print(torch.__version__)
print(torchvision.__version__)
print(torch.cuda.is_available())
print(torch.cuda.device_count())

yaml文件的编写

path: Visdrone  
train: train/images
val: val/images
test: test/images  

nc: 10

names:
  0: pedestrian
  1: people
  2: bicycle
  3: car
  4: van
  5: truck
  6: tricycle
  7: awning-tricycle
  8: bus
  9: motor

不需要指定标注文件，只需要填写图片的路径

数据集的划分（v8）

数据集转换（visdrone为例）

import os
from pathlib import Path
from PIL import Image
from tqdm import tqdm

def visdrone2yolo(dir):
    def convert_box(size, box):
        #Convert VisDrone box to YOLO CxCywh box,坐标进行了归一化
        dw = 1. / size[0]
        dh = 1. / size[1]
        return (box[0] + box[2] / 2) * dw, (box[1] + box[3] / 2) * dh, box[2] * dw, box[3] * dh

    # (dir / 'labels').mkdir(parents=True, exist_ok=True)  # make labels directory
    (dir / 'Annotations_YOLO').mkdir(parents=True, exist_ok=True)  # make labels directory
    pbar = tqdm((dir / 'annotations').glob('*.txt'), desc=f'Converting {dir}')
    for f in pbar:
        img_size = Image.open((dir / 'images' / f.name).with_suffix('.jpg')).size
        lines = []
        with open(f, 'r') as file:  # read annotation.txt
            for row in [x.split(',') for x in file.read().strip().splitlines()]:
                if row[4] == '0':  # VisDrone 'ignored regions' class 0
                    continue
                cls = int(row[5]) - 1
                box = convert_box(img_size, tuple(map(int, row[:4])))
                lines.append(f"{cls} {' '.join(f'{x:.6f}' for x in box)}\n")
                with open(str(f).replace(os.sep + 'annotations' + os.sep, os.sep + 'Annotations_YOLO' + os.sep), 'w') as fl:
                    fl.writelines(lines)  # write label.txt


dir = Path(r'E:\DPL\DeepLearnData\目标检测\航空目标检测数据VisDrone\VisDrone2019') #更改为自己的数据集的路径
# Convert
for d in 'VisDrone2019-DET-train', 'VisDrone2019-DET-val', 'VisDrone2019-DET-test-dev':
    visdrone2yolo(dir / d)  # convert VisDrone annotations to YOLO labels

训练

train.py 作用：用于训练

from ultralytics import YOLO
import os
 
if __name__ == '__main__':
    
    model = YOLO(model="my_lu.yaml")
 
    # Use the model
    results = model.train(data="data.yaml", patience=200,epochs=200, device='0', imgsz=640, batch=4, seed=42)

 用于得出结果

val.py，作用：根据训练好的结果（pt文件），可以运行出训练结果

from ultralytics import YOLO


if __name__ == '__main__':
    model = YOLO('runs/detect/train9/weights/best.pt')  
    metrics = model.val(split='val')  

测试多张图片

test.py 作用：可以用训练完的pt文件检测多张图片，将图片放到yolo/assets文件夹下。

from ultralytics import YOLO

if __name__ == '__main__':
  
    model = YOLO('runs/detect/train9/weights/best.pt') 
    results = model.predict(source="drone-yolo/assets", device='0',save=True)
   
    print(results)

断点训练

将未训练好的pt文件指定路径，和数据集的yaml文件

yolo task=detect mode=train model=runs/detect/train19/weights/last.pt data=data.yaml save=True resume=True

本片文章只是自己的笔记，写的很简单，可能有读者看不懂，望海涵。如有错误，请指教，谢谢。