search

jingwenli0312 / HCVRP_DRL

This repo implements our paper, "Deep Reinforcement Learning for Solving the Heterogeneous Capacitated Vehicle Routing Problem"
133 stars 31 forks source link

Error in eval.py #7

Open eh102 opened 1 year ago

eh102 commented 1 year ago

Hello, I want to run eval.py in HCVRP_DRL. I have followed the input instructions in the official documentation and made sure that the dataset and model are loaded correctly, and the environment matches the documentation.However, I encountered the following error message. Does anyone know how to solve it?

Traceback (most recent call last): File "eval.py", line 226, in eval_dataset(dataset_path, width, opts.softmax_temperature, opts) File "eval.py", line 59, in evaldataset model, = load_model(opts.model, opts.obj) File "/home/ubuntu/Desktop/HCVRP_DRL-main/fleet_v3/utils/functions.py", line 124, in load_model load_data = torch_load_cpu(model_filename) File "/home/ubuntu/Desktop/HCVRP_DRL-main/fleet_v3/utils/functions.py", line 27, in torch_load_cpu return torch.load(load_path, map_location=lambda storage, loc: storage) # Load on CPU File "/home/ubuntu/anaconda3/envs/hcvrp/lib/python3.7/site-packages/torch/serialization.py", line 426, in load return _load(f, map_location, pickle_module, **pickle_load_args) File "/home/ubuntu/anaconda3/envs/hcvrp/lib/python3.7/site-packages/torch/serialization.py", line 613, in _load result = unpickler.load() ModuleNotFoundError: No module named 'nets.attention_model_minsum'

NEWRING commented 1 year ago

Hi, I have met the similar error like you. Have you solved it?

My error message: Traceback (most recent call last): File "...\lib\site-packages\torch\serialization.py", line 713, in load return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args) File "...\lib\site-packages\torch\serialization.py", line 930, in _legacy_load result = unpickler.load() File "...\lib\site-packages\torch\serialization.py", line 746, in find_class return super().find_class(mod_name, name) ModuleNotFoundError: No module named 'problems.hcvrp.problem_hcvrp_minsum'

eh102 commented 1 year ago

I changed my GPU to a 2080ti, modified the settings to the same as the official text Python 3.7, Pytorch 1.3.0. After that it worked fine with my own trained model. So I think it may be a compatibility issue with GPU, CUDA, Python, Pytroch.