search

fangwei123456 / spikingjelly

SpikingJelly is an open-source deep learning framework for Spiking Neural Network (SNN) based on PyTorch.
https://spikingjelly.readthedocs.io
Other
1.22k stars 233 forks source link

Question about DSQN #509

Closed QianpengLi577 closed 3 months ago

QianpengLi577 commented 3 months ago

For faster response

| Reinforcement Learning | @lucifer2859 |

Issue type

SpikingJelly version

0.0.0.0.14

Description

Actually this is not a problem with spikingjelly. I would like to know the specific version number of ptan in the DSQN tutorial. I am using ptan==0.4, but the following error occurs.

File "train.py", line 141, in <module>
    tgt_net = ptan.agent.TargetNet(net, net_)
TypeError: __init__() takes 2 positional arguments but 3 were given

File "utils/atari_wrappers.py", line 32, in reset
    obs, _, done, _ = self.env.step(self.noop_action)
ValueError: too many values to unpack (expected 4)
lucifer2859 commented 3 months ago

Thanks for the feedback, the modified PTAN package is provided in the new version of the code