训练YOLOv5模型时遇runs\train\exp10非目录错误求助

问题描述 #

训练YOLOv5自定义数据时持续报错：

train: weights=yolov5s.pt, cfg=models/yolov5s.yaml, data=data.yaml, hyp=data\hyps\hyp.scratch-low.yaml, epochs=300, batch_size=32, imgsz=640, rect=False, resume=False, nosave=False, noval=False, noautoanchor=False, noplots=False, evolve=None, evolve_population=data\hyps, resume_evolve=None, bucket=, cache=None, image_weights=False, device=, multi_scale=False, single_cls=False, optimizer=SGD, sync_bn=False, workers=8, project=runs\train, name=exp, exist_ok=False, quad=False, cos_lr=False, label_smoothing=0.0, patience=100, freeze=[0], save_period=-1, seed=0, local_rank=-1, entity=None, upload_dataset=False, bbox_interval=-1, artifact_alias=latest, ndjson_console=False, ndjson_file=False
    github: skipping check (not a git repository), for updates see https://github.com/ultralytics/yolov5
    YOLOv5  2024-7-15 Python-3.12.2 torch-2.3.1+cpu CPU
    hyperparameters: lr0=0.01, lrf=0.01, momentum=0.937, weight_decay=0.0005, warmup_epochs=3.0, warmup_momentum=0.8, warmup_bias_lr=0.1, box=0.05, cls=0.5, cls_pw=1.0, obj=1.0, obj_pw=1.0, iou_t=0.2, anchor_t=4.0, fl_gamma=0.0, hsv_h=0.015, hsv_s=0.7, hsv_v=0.4, degrees=0.0, translate=0.1, scale=0.5, shear=0.0, perspective=0.0, flipud=0.0, fliplr=0.5, mosaic=1.0, mixup=0.0, copy_paste=0.0
    TensorBoard: Start with 'tensorboard --logdir runs\train', view at http://localhost:6006/
    Traceback (most recent call last):
      File "C:\Users\Usuário\yolov5_work2024\yolov5-master\train.py", line 986, in <module>
        main(opt)
      File "C:\Users\Usuário\yolov5_work2024\yolov5-master\train.py", line 688, in main
        train(opt.hyp, opt, device, callbacks)
      File "C:\Users\Usuário\yolov5_work2024\yolov5-master\train.py", line 180, in train
        loggers = Loggers(
                  ^^^^^^^^
      File "C:\Users\Usuário\yolov5_work2024\yolov5-master\utils\loggers\__init__.py", line 121, in __init__
        self.tb = SummaryWriter(str(s))
                  ^^^^^^^^^^^^^^^^^^^^^
      File "C:\Users\Usuário\AppData\Local\Programs\Python\Python312\Lib\site-packages\torch\utils\tensorboard\writer.py", line 249, in __init__
        self._get_file_writer()
      File "C:\Users\Usuário\AppData\Local\Programs\Python\Python312\Lib\site-packages\torch\utils\tensorboard\writer.py", line 281, in _get_file_writer
        self.file_writer = FileWriter(
                           ^^^^^^^^^^^
      File "C:\Users\Usuário\AppData\Local\Programs\Python\Python312\Lib\site-packages\torch\utils\tensorboard\writer.py", line 75, in __init__
        self.event_writer = EventFileWriter(**
                            ^^^^^^^^^^^^^^^^
      File "C:\Users\Usuário\AppData\Local\Programs\Python\Python312\Lib\site-packages\tensorboard\summary\writer\event_file_writer.py", line 72, in __init__
        tf.io.gfile.makedirs(logdir)
      File "C:\Users\Usuário\AppData\Local\Programs\Python\Python312\Lib\site-packages\tensorflow\python\lib\io\file_io.py", line 513, in recursive_create_dir_v2
        _pywrap_file_io.RecursivelyCreateDir(compat.path_to_bytes(path))
    tensorflow.python.framework.errors_impl.FailedPreconditionError: runs\train\exp10 is not a directory

已尝试的训练命令：

• 预训练权重训练：

python train.py --img 640 --batch 32 --epochs 300 --data data.yaml --weights yolov5s.pt

• 从头训练：

python train.py --img 640 --batch 32 --epochs 300 --data data.yaml --cfg models/yolov5s.yaml

已查阅相关问题但未找到可行方案。

解决方案 #

• 手动创建日志目录：直接在项目根目录下创建runs\train文件夹，再手动创建报错中指定的子文件夹（如exp10），确保文件夹权限正常，当前用户有读写权限。
• 指定新的日志目录：训练时添加--project参数指定一个全新的目录，绕开已有目录的异常问题：

python train.py --img 640 --batch 32 --epochs 300 --data data.yaml --weights yolov5s.pt --project new_runs/train

• 禁用TensorBoard日志：如果不需要训练过程可视化，添加--noplots参数关闭日志写入：

python train.py --img 640 --batch 32 --epochs 300 --data data.yaml --weights yolov5s.pt --noplots

• 修复路径兼容问题：Windows系统下将命令中的路径分隔符改为正斜杠，或升级TensorFlow到最新版本，解决旧版本的路径处理bug。
• 检查目录权限：确认runs目录及其子目录未被设置为只读，右键文件夹选择“属性”取消只读勾选，确保当前用户拥有读写权限。

内容的提问来源于stack exchange，提问作者Poguinho