Error reported in the section of torch.load('DQN_structure/network_picture/RMFS_DQN_policy_net.pt')

[13261990511]

跑的是DQN算法这是运行结果： Model is controlled by use_NN mode start simulation with DQN algorithm map_xdim: 13 map_ydim: 12 state_number: 3

load NN Traceback (most recent call last): File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\main.py", line 79, in main() File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\main.py", line 61, in main agent = modelController(multi_agv_scene, map_xdim=map_xdim, map_ydim=map_ydim, max_task=max_task, File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\algorithm\DQN_structure\Controller.py", line 34, in init policy_net = torch.load('DQN_structure/network_picture/RMFS_DQN_policy_net.pt') File "E:\conda\Anaconda3\envs\pytorch\lib\site-packages\torch\serialization.py", line 595, in load with _open_file_like(f, 'rb') as opened_file: File "E:\conda\Anaconda3\envs\pytorch\lib\site-packages\torch\serialization.py", line 231, in _open_file_like return _open_file(name_or_buffer, mode) File "E:\conda\Anaconda3\envs\pytorch\lib\site-packages\torch\serialization.py", line 212, in init super(_open_file, self).init(open(name, mode)) FileNotFoundError: [Errno 2] No such file or directory: 'DQN_structure/network_picture/RMFS_DQN_policy_net.pt'

我试着将相对路径改为绝对路径也是一样的报错 然后我尝试着在pycharm上打开RMFS_DQN_policy_net.pt，显示：the file was loaded in a wrong encoding: 'UTF-8', 但是我在打开RMFS_DQN_policy_net_auto.pt时则没有这样的提示

然后我尝试着运行其他强化学习算法，发现其他算法也有类似的问题，比如MADQN： Traceback (most recent call last): File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\main.py", line 79, in main() File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\main.py", line 61, in main agent = modelController(multi_agv_scene, map_xdim=map_xdim, map_ydim=map_ydim, max_task=max_task, File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\algorithm\MADQN_structure\Controller.py", line 35, in init self.create_agent(map_xdim, map_ydim) File "E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\algorithm\MADQN_structure\Controller.py", line 59, in create_agent policy_net = torch.load(os.path.join(self.storage_path, "policy_net.pt")) File "E:\conda\Anaconda3\envs\pytorch\lib\site-packages\torch\serialization.py", line 595, in load with _open_file_like(f, 'rb') as opened_file: File "E:\conda\Anaconda3\envs\pytorch\lib\site-packages\torch\serialization.py", line 231, in _open_file_like return _open_file(name_or_buffer, mode) File "E:\conda\Anaconda3\envs\pytorch\lib\site-packages\torch\serialization.py", line 212, in init super(_open_file, self).init(open(name, mode)) FileNotFoundError: [Errno 2] No such file or directory: 'E:\Reinforcement-learning-for-multi-AGV-pathfinding-main\src\algorithm\MADQN_structure\network_picture\policy_net.pt'

[wangshichao3]

哥们，你弄出来了吗，他这个用训练好的模型为啥还要训练