config file error

[wwxFromTju]

Hey, I try to run the following code: python3 src / main.py --config = categorical_qmix --env-config = sc2 with env_args.map_name = 2s3z and python3 src / main.py --config = tar_qmix --env-config = sc2 with env_args.map_name = 2s3z I found that due to a config error, it cannot be run directly. I modified the runner: "parallel"-> runner: "parallel_x", and modified the representation of the parameters, such as 1e-2-> 0.01

But in the end it still can't run, can you provide complete and runable config file to help others to reproduce?

[TonghanWang]

Hi,

Can you provide more error information?

[wwxFromTju]

Below is the corresponding error message：

[ERROR 10:21:48] pymarl Failed after 0:00:42!
Traceback (most recent calls WITHOUT Sacred internals):
  File "src/main.py", line 35, in my_main
    run(_run, _config, _log)
  File "/home/xxxxxxx/NDQ/src/run.py", line 60, in run
    run_sequential(args=args, logger=logger)
  File "/home/x x x x x x x/NDQ/src/run.py", line 227, in run_sequential
    learner.train(episode_sample, runner.t_env, episode)
  File "/home/xxxxxxx/NDQ/src/learners/categorical_q_learner.py", line 184, in train
    loss.backward()
  File "/home/xxxxxxx/anaconda3/envs/sc_roma/lib/python3.6/site-packages/torch/tensor.py", line 195, in backward
    torch.autograd.backward(self, gradient, retain_graph, create_graph)
  File "/home/xxxxxxx/anaconda3/envs/sc_roma/lib/python3.6/site-packages/torch/autograd/__init__.py", line 99, in backward
    allow_unreachable=True)  # allow_unreachable flag
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.FloatTensor [32, 67, 5, 11]], which is output 0 of SliceBackward, is at version 1; expected version 0 instead. Hint: enable anomaly detection to find the operation that failed to compute its gradient, with torch.autograd.set_detect_anomaly(True).

[TonghanWang]

It seems that this is not a config file error.

[TonghanWang]

What is the version of your pytorch?

[wwxFromTju]

Changed to 0.4, it looks like it can run for a while. Thanks for the quick response.

I still suggest that you modify your current config file, the problems are as I mentioned above

[TonghanWang]

We have updated the README, including the detailed command to reproduce the results in our paper.

Thanks for your comments. Feel free to contact us if you have any other questions.

[wwxFromTju]

Thank you for your very quick reply and great work.