YOLOV5——将图片和标注数据按比例切分为训练集和测试集

将图片和标注数据按比例切分后存储至新的路径下

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

# 将图片和标注数据按比例切分为 训练集和测试集   import os from shutil import copy2 # 原始路径 image_original_path = "../image_data/seed/images/" label_original_path = "../image_data/seed/labels/" # 上级目录 parent_path = os.path.dirname(os.getcwd()) # 训练集路径 train_image_path = os.path.join(parent_path, "image_data/seed/train/images/") train_label_path = os.path.join(parent_path, "image_data/seed/train/labels/") # 测试集路径 test_image_path = os.path.join(parent_path, 'image_data/seed/test/images/') test_label_path = os.path.join(parent_path, 'image_data/seed/test/labels/')   # 检查文件夹是否存在 def mkdir():     if not os.path.exists(train_image_path):         os.makedirs(train_image_path)     if not os.path.exists(train_label_path):         os.makedirs(train_label_path)       if not os.path.exists(test_image_path):         os.makedirs(test_image_path)     if not os.path.exists(test_label_path):         os.makedirs(test_label_path)     def main():     mkdir()     # 复制移动图片数据     all_image = os.listdir(image_original_path)     for i in range(len(all_image)):         if i % 10 != 0:             copy2(os.path.join(image_original_path, all_image[i]), train_image_path)         else:             copy2(os.path.join(image_original_path, all_image[i]), test_image_path)       # 复制移动标注数据     all_label = os.listdir(label_original_path)     for i in range(len(all_label)):         if i % 10 != 0:             copy2(os.path.join(label_original_path, all_label[i]), train_label_path)         else:             copy2(os.path.join(label_original_path, all_label[i]), test_label_path)     if __name__ == '__main__':     main()