sasaffcxzc 2022-09-09 13:24 采纳率: 65.5%
浏览 172
已结题

请教yolov6训练报错


Training start...
C:\Users\11754\.conda\envs\yoyo\lib\site-packages\torch\cuda\amp\grad_scaler.py:115: UserWarning: torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.
  warnings.warn("torch.cuda.amp.GradScaler is enabled, but CUDA is not available.  Disabling.")
ERROR in training loop or eval/save model.

Training completed in 0.000 hours.
Traceback (most recent call last):
  File "J:\dll\yolo6\YOLOv6-0.2.0\yolov6\core\engine.py", line 97, in train
    self.train_before_loop()
  File "J:\dll\yolo6\YOLOv6-0.2.0\yolov6\core\engine.py", line 226, in train_before_loop
    distill_weight = self.cfg.model.head.distill_weight,
  File "J:\dll\yolo6\YOLOv6-0.2.0\yolov6\utils\config.py", line 30, in __getattr__
    raise ex
AttributeError: 'ConfigDict' object has no attribute 'distill_weight'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "J:\dll\yolo6\YOLOv6-0.2.0\tools\train.py", line 126, in <module>
    main(args)
  File "J:\dll\yolo6\YOLOv6-0.2.0\tools\train.py", line 116, in main
    trainer.train()
  File "J:\dll\yolo6\YOLOv6-0.2.0\yolov6\core\engine.py", line 105, in train
    self.train_after_loop()
  File "J:\dll\yolo6\YOLOv6-0.2.0\yolov6\core\engine.py", line 261, in train_after_loop
    strip_optimizer(save_ckpt_dir, self.epoch)  # strip optimizers for saved pt model
AttributeError: 'Trainer' object has no attribute 'epoch'

AttributeError: 'ConfigDict' object has no attribute 'distill_weight'

During handling of the above exception, another exception occurred:

AttributeError: 'Trainer' object has no attribute 'epoch' 一直提示这个 我都pip安装了运行还是提示 不知道哪点问题

  • 写回答

2条回答 默认 最新

查看更多回答(1条)

报告相同问题?

问题事件

  • 系统已结题 9月17日
  • 已采纳回答 9月9日
  • 修改了问题 9月9日
  • 创建了问题 9月9日

悬赏问题

  • ¥15 用visual studi code完成html页面
  • ¥15 聚类分析或者python进行数据分析
  • ¥15 逻辑谓词和消解原理的运用
  • ¥15 三菱伺服电机按启动按钮有使能但不动作
  • ¥15 js,页面2返回页面1时定位进入的设备
  • ¥50 导入文件到网吧的电脑并且在重启之后不会被恢复
  • ¥15 (希望可以解决问题)ma和mb文件无法正常打开,打开后是空白,但是有正常内存占用,但可以在打开Maya应用程序后打开场景ma和mb格式。
  • ¥20 ML307A在使用AT命令连接EMQX平台的MQTT时被拒绝
  • ¥20 腾讯企业邮箱邮件可以恢复么
  • ¥15 有人知道怎么将自己的迁移策略布到edgecloudsim上使用吗?