使用paddlex训练图片多标签识别报错：NotADirectoryError: [WinError 267] 目录名称无效

[syge]

Checklist:

• 查找历史相关issue寻求解答
• 翻阅FAQ
• 翻阅PaddleX 文档
• 确认bug是否在新版本里还未修复

描述问题

复现

1. 您是否已经正常运行我们提供的教程？

2. 您是否在教程的基础上修改代码内容？还请您提供运行的代码

3. 您使用的数据集是？

4. 请提供您出现的报错信息及相关log

环境

1. 请提供您使用的PaddlePaddle和PaddleX的版本号

2. 请提供您使用的操作系统信息，如Linux/Windows/MacOS

3. 请问您使用的Python版本是？

4. 请问您使用的CUDA/cuDNN的版本号是？

使用python版本是：3.10.14

跑的就是提供的示例：python main.py -c paddlex/configs/multilabel_classification/PP-LCNet_x1_0_ML.yaml -o Global.mode=train -o Global.dataset_dir=./dataset/mlcls_nus_examples

配置文件信息：
Global:
model: PP-LCNet_x1_0_ML

mode: check_dataset

mode: train
dataset_dir: dataset/mlcls_nus_examples
device: gpu:0
output: "output"

CheckDataset:
convert:
enable: False
src_dataset_type: null
split:
enable: False
train_percent: null
val_percent: null

Train:
num_classes: 33
epochs_iters: 20
batch_size: 32
learning_rate: 0.1
pretrain_weight_path: https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LCNet_x1_0_ML_pretrained.pdparams
warmup_steps: 5
resume_path: null
log_interval: 20
eval_interval: 1
save_interval: 1

Evaluate:
weight_path: "output/best_model/best_model.pdparams"
log_interval: 1

Export:
weight_path: https://paddle-model-ecology.bj.bcebos.com/paddlex/official_pretrained_model/PP-LCNet_x1_0_ML_pretrained.pdparams

Predict:
batch_size: 1
model_dir: "output/best_model/inference"
input: "https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_image_classification_001.jpg"
kernel_option:
run_mode: paddle

具体报错信息是：
Traceback (most recent call last):
File "D:\python-application\PaddleX\paddlex\utils\result_saver.py", line 29, in wrap
result = func(self, *args, **kwargs)
File "D:\python-application\PaddleX\paddlex\engine.py", line 41, in run
self._model.train()
File "D:\python-application\PaddleX\paddlex\model.py", line 94, in train
trainer.train()
File "D:\python-application\PaddleX\paddlex\modules\base\trainer.py", line 71, in train
train_result = self.pdx_model.train(**train_args)
File "D:\python-application\PaddleX\paddlex\repo_apis\PaddleClas_api\cls\model.py", line 123, in train
return self.runner.train(
File "D:\python-application\PaddleX\paddlex\repo_apis\PaddleClas_api\cls\runner.py", line 50, in train
return self.run_cmd(
File "D:\python-application\PaddleX\paddlex\repo_apis\base\runner.py", line 344, in run_cmd
retcode = asyncio.run(
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\runners.py", line 44, in run
return loop.run_until_complete(main)
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\base_events.py", line 649, in run_until_complete
return future.result()
File "D:\python-application\PaddleX\paddlex\repo_apis\base\runner.py", line 308, in _tee_proc_call
proc = await proc_call
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\subprocess.py", line 218, in create_subprocess_exec
transport, protocol = await loop.subprocess_exec(
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\base_events.py", line 1681, in subprocess_exec
transport = await self._make_subprocess_transport(
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\windows_events.py", line 399, in _make_subprocess_transport
transp = _WindowsSubprocessTransport(self, protocol, args, shell,
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\base_subprocess.py", line 36, in init
self._start(args=args, shell=shell, stdin=stdin, stdout=stdout,
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\windows_events.py", line 901, in _start
File "D:\software\aconda\envs\dbgpt_env\lib\asyncio\windows_utils.py", line 153, in init
super().init(args, stdin=stdin_rfd, stdout=stdout_wfd,
File "D:\software\aconda\envs\dbgpt_env\lib\subprocess.py", line 971, in init
File "D:\software\aconda\envs\dbgpt_env\lib\subprocess.py", line 1456, in _execute_child
hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
NotADirectoryError: [WinError 267] 目录名称无效。
Unknown option: -e

[changdazhou]

尝试将dataset_dir的路径修改成windows目录表达形式看看

[syge]

改成使用绝对路径，依然报上述错误
python main.py -c paddlex/configs/multilabel_classification/PP-LCNet_x1_0_ML.yaml -o Global.mode=train -o Global.dataset_dir=D:\python-application\PaddleX\dataset\mlcls_nus_examples

[changdazhou]

这样试一下呢：D:\python-application\PaddleX\dataset\mlcls_nus_examples

[TingquanGao]

The issue has no response for a long time and will be closed. You can reopen or new another issue if are still confused.

---

From Bot

[renchengwei]

改成使用绝对路径，依然报上述错误 python main.py -c paddlex/configs/multilabel_classification/PP-LCNet_x1_0_ML.yaml -o Global.mode=train -o Global.dataset_dir=D:\python-application\PaddleX\dataset\mlcls_nus_examples

请问您最终解决这个问题了吗？我现在也遇到了这个问题，请教您一下，谢谢！