Yuanhao Geng
Yuanhao Geng
> > > You probably don't want to have _both_ `hs=1` and `negative=10` – both modes enabled – at once. > > > > > > Why not? Isnt it...
For each item, the same interval close form is expected. So now we set the lowest edge as `min - 0.1% * (max - min)` when close is right to...
When you use `set_levels`, pandas directly change the old level to a new one without aligning the original data value. Is this the desired result? @d-s-dc
You can use `--max-epoch` to specify the number of epoches. Since Graphormer is based on Fairseq now, more training options can be found in [FairSeq's DOC](https://fairseq.readthedocs.io/en/latest/command_line_tools.html).
Hello, whether ``i!=j`` still be needed?
easy-ros2arm好像也是负责人？不建议两个同时开
这个作者b站视频的[网站](https://imitation-learning-blog.github.io/about/)里用的是No Derivatives的协议，使用相关材料整理是否获得作者许可？
我举个第二章的例子，比如开头这里这三个公式是视频搬过来的，但是根据描述并没有把这个意义描述的比较清楚，比如红框的里几个符号的含义至少需要说明，这个补充的部分属于规划的还未完成的30%之内吗  