Daochen Zha
Daochen Zha
@DoxakisCh Thanks for the question! Yes, you are correct. The implementation is chance sampling CFR. The deck is randomly dealt for each new game for traversal. We will modify the...
@FranzPfeifroth To my experience, whether it will work mainly depend on the feature designs. My suggestion is to mainly base on DMC in https://github.com/datamllab/rlcard/blob/master/examples/run_dmc.py, which supports multiple actors so the...
@sword134 Thanks for the proposal! Our goal of blackjack is a simple example to get started. Thus, we make it as simple as possible (in the current version, we have...
@LizChen2000 Hey, thanks for the feedback. Could you let me which file it is? We don't have a file named `train_nn.py`.
@lssr Calculating this number is very challenging. An estimation is here https://github.com/datamllab/rlcard#available-environments Some are from other sources, and some are estimated by us. However, correctness is not guaranteed. The total...
@billh0420 I feel like it is not that easy. There are two aspects we may need to consider. 1. the cards on the board, since the cards on the board...
@billh0420 @justinkterry Thanks a lot! The calculation looks reasonable to me. I will put it into README later. We have also 2 other dimensions. One is action size. The other...
@RioMarPa Hi, could you provide more details? Like what errors you have met?
@RioMarPa Have you set any breakpoints in Visual Studio?
@RioMarPa Have tried directly run the script in the terminal? Like `python3 XXX.py`