Windows系统测试用例时出现下面问题 请问应该如何解决？

[2778817374]

测试代码：
import xuance
runner = xuance.get_runner(method='dqn',
env='classic_control',
env_id='CartPole-v1',
is_test=False)
runner.run()
报错：
test_dqn.py:None (test_dqn.py)
test_dqn.py:6: in
runner.run()
xuance\torch\runners\runner_drl.py:93: in run
self.agent.save_model("final_train_model.pth")
xuance\torch\agents\agent.py:90: in save_model
self.learner.save_model(model_path)
xuance\torch\learners\learner.py:25: in save_model
torch.save(self.policy.state_dict(), model_path)
C:\Users\G11.conda\envs\xuance_env\lib\site-packages\torch\serialization.py:422: in save
with _open_zipfile_writer(f) as opened_zipfile:
C:\Users\G11.conda\envs\xuance_env\lib\site-packages\torch\serialization.py:309: in _open_zipfile_writer
return container(name_or_buffer)
C:\Users\G11.conda\envs\xuance_env\lib\site-packages\torch\serialization.py:287: in init
super(_open_zipfile_writer_file, self).init(torch._C.PyTorchFileWriter(str(name)))
E RuntimeError: Parent directory E:\JYC_CODE\强化学习\xuance./models/dqn does not exist.
collected 0 items / 1 error

是由于Windows系统下文件路径和Linux文件路径不一致导致的么？请问应该如何解决？谢谢~

[wenzhangliu]

你好，方便让我看一下torch和xuance的版本号吗？先排除一下是不是版本问题

[2778817374]

谢谢您，
torch==1.13.0
xuance=='v1.0.9'
xuance的版本是在xuance库中的__init__.py中查到的，您看这种查法是正确的么？

[2778817374]

此外，在运行demo_marl.py测试时，我将parser.add_argument("--device", type=str, default="cuda:0")修改为
parser.add_argument("--device", type=str, default="cpu"),
代码运行时报错：
Traceback (most recent call last):
File "E:\JYC_CODE\xuance\demo_marl.py", line 22, in
is_test=parser.test)
File "E:\JYC_CODE\xuance\xuance\common\common_tools.py", line 241, in get_runner
runner = run_REGISTRYargs[0].runner if type(args) == list else run_REGISTRYargs.runner
File "E:\JYC_CODE\xuance\xuance\torch\runners\runner_pettingzoo.py", line 101, in init
self.marl_agents.append(REGISTRY_Agent[arg.agent](arg, self.envs, arg.device))
File "E:\JYC_CODE\xuance\xuance\torch\agents\multi_agent_rl\maddpg_agents.py", line 21, in init
policy = REGISTRY_Policyconfig.policy
File "E:\JYC_CODE\xuance\xuance\torch\policies\deterministic_marl.py", line 499, in init
normalize, initialize, activation, device)
File "E:\JYC_CODE\xuance\xuance\torch\policies\deterministic_marl.py", line 447, in init
actor_hidden_size, normalize, initialize, activation, device)
File "E:\JYC_CODE\xuance\xuance\torch\policies\deterministic_marl.py", line 393, in init
layers.extend(mlp_block(input_shape[0], action_dim, None, nn.Tanh, initialize, device)[0])
File "E:\JYC_CODE\xuance\xuance\torch\utils\layers.py", line 15, in mlp_block
linear = nn.Linear(input_dim, output_dim, device=device)
File "C:\Users\G11.conda\envs\xuance_env\lib\site-packages\torch\nn\modules\linear.py", line 96, in init
self.weight = Parameter(torch.empty((out_features, in_features), **factory_kwargs))
TypeError: empty() received an invalid combination of arguments - got (tuple, dtype=NoneType, device=str), but expected one of:

• (tuple of ints size, *, tuple of names names, torch.memory_format memory_format, torch.dtype dtype, torch.layout layout, torch.device device, bool pin_memory, bool requires_grad)
• (tuple of SymInts size, *, torch.memory_format memory_format, Tensor out, torch.dtype dtype, torch.layout layout, torch.device device, bool pin_memory, bool requires_grad)
  您看应该如何解决？
  此外，我看代码中torch的版本为1.13.0，如果我要更改torch为GPU版本，您看应该修改torch-gpu版本为多少合适，是否要和我本地原来装的CUDA和CUDNN版本相对应，我的cuda版本为12.0.140，是否要对CUDA和cudnn进行降级，麻烦您了，谢谢！

[wenzhangliu]

您好，第二个问题和torch版本无关，是由于我们近期优化MARL部分代码出现的小bug，目前已修复并上传最新代码了。给您带来不便非常抱歉。

但是第一个问题在我电脑上没有发生过，我用的是MacOS，我跟您一样也怀疑是Windows识别路径方式和Linux不一样导致的。感谢您提出这个问题，我们会在Windows上充分测试并排除类似问题。

[2778817374]

万分感谢！！！