-
- [ ] I have marked all applicable categories:
+ [ ] exception-raising bug
+ [ ] RL algorithm bug
+ [ ] documentation request (i.e. "X is missing from the documentation.")
+ [x] ne…
-
### What is the problem?
When trying to train with a DQN (or really any other algorithm - I even tried it with PPO) with the latest nightly installation of Ray , I seem to get a weird error throw…
-
I was running this project: `https://github.com/transedward/pytorch-dqn`
`python main.py`
and get `Segmentation fault (core dumped)`
The python version is 3.5, pytorch is bulit from source (s…
-
**Describe the bug**
Hi.
I was testing DQN using pytorch and use env = UnityEnvironment(filename=None) connect to the GridWorld environment.
After each episode end, the script would call env.rese…
-
As the title says.
-
Hi @lanctot , first of all this is amazing collaborative effort! Thanks for this library.
I tried training DQN with Solitaire in Pytorch by tweaking the code [here](https://github.com/deepmind/open_s…
-
#### What is your question?
I just implemented a customized LSTM model on fairseq and the training process has no bug. Then in the generation process, I could only successfully generate the first b…
-
Hi Petros,
in your SAC_discrete code you are using the following in `SAC_Discrete.py`:
```
min_qf_next_target = action_probabilities * (torch.min(qf1_next_target, qf2_next_target) - self.alpha * …
-
### Question
The model.learn function in SB3 is to generate one action according to the state and then obtain the reward, then the model is trained. However, if I have multiple state-action-rewar…
-
Dqns should be added to the reinforcement learning. What do you think @soumith ?