Update to run on torch 0.4

higgsfield / RL-Adventure

Pytorch Implementation of DQN / DDQN / Prioritized replay/ noisy networks/ distributional values/ Rainbow/ hierarchical RL

2.99k stars 587 forks source link

Open tsaoyu opened 5 years ago

tsaoyu commented 5 years ago

Trying to address two issues:

return action as an item to fix runtime error
```
UserWarning: volatile was removed and now has no effect. Use `with torch.no_grad():` instead.
```
may related to: https://github.com/SeanNaren/deepspeech.pytorch/issues/277
change number of actions from global to a local one which I believe that was your intention