paddlecls特征提取模型微调

[chinesejunzai12]

1. 相关版本如下:
   paddleclas 2.5.1
   paddlepaddle 2.5.2
   paddlepaddle-gpu 2.5.2.post112
   2.python版本号: Python 3.7.13
2. cuda/cudnn版本 cuda11.2/cudnn8.1
   文档链接: https://github.com/PaddlePaddle/PaddleClas/blob/release/2.5/docs/zh_CN/training/metric_learning/training.md
   在运行命令之后python3.7 tools/train.py
   -c ./ppcls/configs/quick_start/MobileNetV1_retrieval.yaml
   -o Arch.Backbone.pretrained=True
   -o Global.device=gpu 报如下问题:
   Traceback (most recent call last):
   File "tools/train.py", line 32, in
   engine = Engine(config, mode="train")
   File "/mnt/project/PaddleClas/ppcls/engine/engine.py", line 99, in init
   self.device = paddle.set_device(self.config["Global"]["device"])
   File "/usr/local/lib/python3.7/dist-packages/paddle/device/init.py", line 266, in set_device
   place = _convert_to_place(device)
   File "/usr/local/lib/python3.7/dist-packages/paddle/device/init.py", line 181, in _convert_to_place
   "The device should not be 'gpu', "
   ValueError: The device should not be 'gpu', since PaddlePaddle is not compiled with CUDA
   请问这个bug是因为版本不匹配还是其他问题?

[cuicheng01]

请问您装的是Paddle的GPU版本吗？建议重新uninstall后再装下GPU版的Paddle呢

[chinesejunzai12]

是paddle的GPU版本, 已经解决了, 是版本的问题