TypeError: unbound method _TensorBase.detach() needs an argument

[Andrei-Iosifescu123]

Hi, I get this error when I try to train it with python3 main.py train --save_path /home/andrei/td-gammon/td_gammon/saved_models/exp1 --save_step 10000 --episodes 100000 --name exp1 --type nn --lr 0.1 --hidden_units 40. This is the full output:

Traceback (most recent call last):
  File "/home/andrei/td-gammon/td_gammon/main.py", line 68, in <module>
    args.func(args)
  File "/home/andrei/td-gammon/td_gammon/utils.py", line 72, in args_train
    net.train_agent(env=env, n_episodes=n_episodes, save_path=save_path, save_step=save_step, eligibility=eligibility, name_experiment=name)
  File "/home/andrei/td-gammon/td_gammon/model.py", line 91, in train_agent
    action = agent.choose_best_action(actions, env)
  File "/home/andrei/td-gammon/td_gammon/agents.py", line 80, in choose_best_action
    best_action_index = int(np.argmax(values)) if self.color == WHITE else int(np.argmin(values))
  File "<__array_function__ internals>", line 200, in argmin
  File "/home/andrei/.local/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 1338, in argmin
    return _wrapfunc(a, 'argmin', axis=axis, out=out, **kwds)
  File "/home/andrei/.local/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 54, in _wrapfunc
    return _wrapit(obj, method, *args, **kwds)
  File "/home/andrei/.local/lib/python3.10/site-packages/numpy/core/fromnumeric.py", line 43, in _wrapit
    result = getattr(asarray(obj), method)(*args, **kwds)
  File "/home/andrei/.local/lib/python3.10/site-packages/torch/_tensor.py", line 956, in __array__
    return Tensor.detach().numpy()
TypeError: unbound method _TensorBase.detach() needs an argument