tianshou
tianshou copied to clipboard
thu-ml
An elegant PyTorch deep reinforcement learning library.
- [ ] I have marked all applicable categories: + [ ] exception-raising bug + [ ] RL algorithm bug + [ ] documentation request (i.e. "X is missing from...
- I have visited the [source website](https://github.com/thu-ml/tianshou/) - I have searched through the [issue tracker](https://github.com/thu-ml/tianshou/issues) for duplicates - I have mentioned version numbers, operating system and environment, where applicable: ```python...
- [ ] I have marked all applicable categories: + [ ] exception-raising bug + [ ] RL algorithm bug + [ ] documentation request (i.e. "X is missing from...
Hello, I have a question when using a `gym.ObservationWrapper` for training and testing as a mask. There are many actions in my env and some of them are unavailable in...
In most multi-agent scenarios, e.g., sc2, dota2, agents should execute actions simultaneously instead of step by step. So the parallel environment where the step function receives actions of all ready...
https://github.com/thu-ml/tianshou/blob/f13e415eb0de55baca5dc0d6fae39d6a38e8bc0b/tianshou/policy/modelfree/dqn.py#L167 It seems `state` is not used during training even when specifying a recurrent net. Am I missing something, or is it expected?
- [x] I have marked all applicable categories: + [ ] exception-raising fix + [ ] algorithm implementation fix + [ ] documentation modification + [x] new feature - [x]...
I want to sample multiple actions from the network's raw output, so I canceled the softmax option from original network and divide the output vector into N equal portions. Then...
- [ ] I have marked all applicable categories: + [ ] exception-raising bug + [ ] RL algorithm bug + [ ] documentation request (i.e. "X is missing from...
- [x] I have marked all applicable categories: + [ ] exception-raising bug + [ ] RL algorithm bug + [ ] documentation request (i.e. "X is missing from the...
