深度学习|Yolov5训练指南—CoCo格式数据集


Yolov5训练指南—CoCo格式数据集

  • 1 准备工作
  • 2 将coco数据集转换为yolo数据集
  • 3 训练参数定义
  • 4 训练模型
  • 5 预测

1 准备工作
  1. 【深度学习|Yolov5训练指南—CoCo格式数据集】训练Yolo模型要准备的文件及文件格式如下:
    /trianing # 根目录 /datasets # 数据集目录(可以任意取名) /images /train /val /labels /train /val /yolov5

  2. 先创建一个training文件夹mkdir training/
  3. 在training文件夹下使用git clone把yolov5克隆下来并安装依赖
    cd training
    git clone clone https://github.com/ultralytics/yolov5
    pip install -qr requirements.txt
  4. 检查pytorch和torchvision的版本
    pip install --upgrade torch
    pip install --upgrade torchvision
  5. 检查label是否连续,如不连续需要重新编码
  6. 使用Weights & Bias进行可视化,其中login的API可以在Weights & Biases上获取。
    %load_ext tensorboard
    %tensorboard --logdir /kaggle/training/yolov5/runs
    %pip install -q --upgrade wandb
    import wandb wandb.login()

2 将coco数据集转换为yolo数据集
  1. 使用json.load(open(file_path,'r'))读取数据
  2. 创建一个csv存放图片的id和文件名
  3. 读取2创建的csv,用train_test_split来切分训练集和验证集
  4. 在切分出来的trian和test文件中分别新增一列,用来标记该图片为训练图片还是测试图片
    train['split']='train'
    val['split']='val'
    df = pd.concat([trian,val],axis=0).rest_index(drop=True)
  5. 将每张图片的标签单独存放到各自的.txt文件中,其中coco数据集的annotation是[lowest_x,lowest_y,w,h],而yolo的annotation要求[center_x,center_y,w,h],使用以下函数:
    def coco2yolo(image_w,image_h,annotation): """Convert coco format data into yolo format data. Note: x,y in coco format are lowest left x and y. x,y in yolo format are center x,y. """ x,y,w,h = annotation['bbox']x = (x+w)/2.0 y = (y+h)/2.0return (x/image_w,y/image_h,w/image_w,h/image_h)

  6. 在training目录下创建dataset文件夹
    os.makedirs('/kaggle/training/cowboy/images/train', exist_ok=True)
    os.makedirs('/kaggle/training/cowboy/images/test', exist_ok=True)
    os.makedirs('/kaggle/training/cowboy/labels/train', exist_ok=True)
    os.makedirs('/kaggle/training/cowboy/labels/test', exist_ok=True)
  7. 将对应的图片和标签复制到train和test文件夹下
  8. 创建一个.ymal文件,该文件用于存放:
    1)训练数据和测试数据的路径
    2)类别总数
    3)类别对应的名称
    data_yaml = dict(train='/kaggle/training/cowboy/images/train/' ,val='/kaggle/training/cowboy/images/test/' ,nc=5 ,names=['belt', 'sunglasses', 'boot', 'cowboy_hat', 'jacket']) with open('/kaggle/training/yolov5/data/data.yaml', 'w') as outfile: yaml.dump(data_yaml, outfile, default_flow_style=True)

3 训练参数定义 参数如下:
lr0: 0.01# initial learning rate (SGD=1E-2, Adam=1E-3) lrf: 0.2# final OneCycleLR learning rate (lr0 * lrf) momentum: 0.937# SGD momentum/Adam beta1 weight_decay: 0.0005# optimizer weight decay 5e-4 warmup_epochs: 3.0# warmup epochs (fractions ok) warmup_momentum: 0.8# warmup initial momentum warmup_bias_lr: 0.1# warmup initial bias lr box: 0.05# box loss gain cls: 0.5# cls loss gain cls_pw: 1.0# cls BCELoss positive_weight obj: 1.0# obj loss gain (scale with pixels) obj_pw: 1.0# obj BCELoss positive_weight iou_t: 0.20# IoU training threshold anchor_t: 4.0# anchor-multiple threshold # anchors: 3# anchors per output layer (0 to ignore) fl_gamma: 0.0# focal loss gamma (efficientDet default gamma=1.5) hsv_h: 0.015# image HSV-Hue augmentation (fraction) hsv_s: 0.7# image HSV-Saturation augmentation (fraction) hsv_v: 0.4# image HSV-Value augmentation (fraction) degrees: 0.0# image rotation (+/- deg) translate: 0.1# image translation (+/- fraction) scale: 0.5# image scale (+/- gain) shear: 0.0# image shear (+/- deg) perspective: 0.0# image perspective (+/- fraction), range 0-0.001 flipud: 0.0# image flip up-down (probability) fliplr: 0.5# image flip left-right (probability) mosaic: 1.0# image mosaic (probability) mixup: 0.0# image mixup (probability) copy_paste: 0.0# segment copy-paste (probability)

4 训练模型 选择模型的时候,可以选择不同大小的模型:Yolov5
BATCH_SIZE = 32 # wisely choose, use the largest size that can feed up all your gpu ram EPOCHS = 5 MODEL = 'yolov5m.pt'# 5s, 5m 5l name = f'{MODEL}_BS_{BATCH_SIZE}_EP_{EPOCHS}'# 在yolov5目录下 !python train.py --batch {BATCH_SIZE} \ --epochs {EPOCHS} \ --data data.yaml \ --weights {MODEL} \ --save-period 1 \ --project /kaggle/working/kaggle-cwoboy \ --name {name} \ -- workers 4

5 预测
  1. 训练好的模型存放在W&B中,把最好的模型下载下来并上传到kaggle
  2. 将测试图片放到VALID_PATH文件夹下
  3. 回到yolov5路径下跑下面这行代码进行预测
    !python detect.py --weights {MODEL_PATH} \ --source {VALID_PATH} \ --conf 0.546 \ --iou-thres 0.5 \ --save-txt \ --save-conf \ --augment

最终预测结果在/kaggle/training/yolov5/runs/detect/exp/labels/
  1. 若要转换成coco的坐标使用下面这个函数
    def yolo2cc_bbox(img_width, img_height, bbox): x = (bbox[0] - bbox[2] * 0.5) * img_width y = (bbox[1] - bbox[3] * 0.5) * img_height w = bbox[2] * img_width h = bbox[3] * img_heightreturn (x, y, w, h)

  2. 若之前对标签进行了编码,要把标签再映射回去
  3. 若要可视化结果可以使用Opencv或PIL读取yolov5/runs/detect/exp/下的照片。
    m = Image.open('/kaggle/training/yolov5/runs/detect/exp/0007c3f55f707547.jpg') im

    深度学习|Yolov5训练指南—CoCo格式数据集
    文章图片

    推荐阅读