深度碎片

Thanks, I am good. I am looking forward to reading your neural translation model!

 在视频一开始就讲到了这一点。A 是agent's action, 由agent生成；R是由Environment生成

关于问题1，针对这个例子，无法再举例了，需要有更详细的例子来对比讲解； 关于问题2， Env生成的Rewards都不是随机的，但Rewards 值被生成，参数调整在”大脑“也就是模型参数中发生

How to use R in NetLogo? [how to install R extension](https://github.com/NetLogo/R-Extension#installing-r) [how to configure R extension](https://github.com/NetLogo/R-Extension#installing-r) [the tweak to note video](url) [a solution issue](https://github.com/NetLogo/R-Extension/issues/8#issuecomment-338238587)

互动可视化的为你解读 `r'/([^/]+)_\d+.jpg`, 请前往regexr.com/48vci --- ## 人工解读 **`pat = r'/([^/]+)_\d+.jpg$'`** * `$` = 搜索从这里开始 * `.jpg` = 一摸一样的字符 * `\d` = 数字, `\d+` = 任意多数字 * `_` = 一摸一样的字符 * `()`...

对比我的笔记视频看看，相关细节应该都有讲到；如果还是有疑问，把英文和中文视频中出现疑问的时间点（如果中文视频也没有解决你的疑问）标注清楚，同时告诉我为什么中文视频没有解决你的疑惑；我会尝试回答

Thanks Dan, I will try to participate on those links you suggested later. Since I am working on [your examples](https://github.com/shiffman/LearningProcessing) on processing3, if I found a trivial issue to correct...

Thanks Dan, I will keep an eye on your repos and do my best to assist you on those issues which I can manage.

Thanks for asking the question which I want to ask for a while. I love polars, and I would love it even more if it could utilize GPU!

### notebook from scratch - rewrite a notebook as you tell a story