Yuchen Shi

[question] Mask Actions

1

Hi, I'm wondering if there is a way to mask actions? To be specific, what I need is to customize the decoder with a mask input so that its output...

ImportError: /../lib/libstdc++.so.6: version `GLIBCXX_3.4.30' not found

6

Environment: - System: WSL, Ubuntu 22.04.2 LTS - Python 3.7.17, 3.8.18, 3.9.18 Reproduce: ``` conda create -n football python=3.8 conda activate football ``` ``` sudo apt-get install git cmake build-essential...

course2中的alphabeta-reversi和mcts-reversi性能不佳, 提交后胜率在0.4左右, 可能是因为board采用了浅拷贝, 比如把[alphabeta-reversi](https://github.com/jidiai/SummerCourse2022/blob/d9cd874cd33276c7ac5244ec7f68943452a205a8/course2/examples/alphabeta-reversi/submission.py#L52)修改为`newBoard = copy.deepcopy(board)`后胜率就可达0.8, 或者也可以把board用np.array存储也是一样的效果.

Hi, would you mind explain a little bit more about equation 11 in the paper? I have $$ \begin{aligned} \nabla_\mathbf{X} \mathcal{H}(\mathbf{X}, M) &= \nabla_{\mathbf{X}}\frac{\mathcal{C}_M(\mathbf{X})-\mathcal{C}_{M^{\prime}}(\mathbf{X})}{\mathcal{C}_{M^{\prime}}(\mathbf{X})}\\ &=\frac{1}{\mathcal{C}_{M^\prime}(\mathbf{X})}\left[-\mathcal{C}_M(\mathbf{X})\nabla_{\mathbf{X}}\log C_{M^\prime}(\mathbf{X})+\nabla_{\mathbf{X}}\mathcal{C}_M (\mathbf{X})\right] \end{aligned} $$ but...

**Describe the bug** When the taskbar is on the right or bottom, everything works as expected. But when the taskbar is on the left or top, the mouse's highlight is...

### Question Suppose the Dict action space is like: ```python spaces.Dict({ "discrete": spaces.Discrete(2), "multibinary": spaces.MultiBinary(3), }) ``` After reading [Tutorial: Action Masking - PettingZoo Docs](https://pettingzoo.farama.org/tutorials/custom_environment/3-action-masking), I guess the corresponding action...

zh-cn date format not localized

1

- hugo version: extended_0.152.2 - stack version: v3.31.0 The date shown in https://dev.stack.jimmycai.com/ar is correctly localized as `مارس 09, 2019`. However, the date in https://dev.stack.jimmycai.com/zh-cn is shown as `Sep 09,...

According to LKH's `TYPE` conventions (e.g., see [LKH3/SRC/ReadProblem.c](https://github.com/c4v4/LKH3/blob/5bd63680970445d4a595ca2c966ac593b0b32ad1/SRC/ReadProblem.c#L88)): ``` * CEIL_2D Weights are Euclidean distances in 2-D rounded up * CEIL_3D Weights are Euclidean distances in 3-D rounded up *...