浏览代码

Added information about input normalization to the best practices doc (#220)

/tag-0.2.1
Arthur Juliani 7 年前
当前提交
22d9e291
共有 2 个文件被更改,包括 67 次插入1 次删除
  1. 4
      docs/best-practices.md
  2. 64
      images/normalization.png

4
docs/best-practices.md


## States
* States should include all variables relevant to allowing the agent to take the optimally informed decision.
* Categorical state variables such as type of object (Sword, Shield, Bow) should be encoded in one-hot fashion (ie `3` -> `0, 0, 1`).
* Rotation information on GameObjects should be recorded as `state.Add(transform.rotation.eulerAngles.y/180.0f-1.0f);` rather than `state.Add(transform.rotation.y);`.
* Besides encoding non-numeric values, all inputs should be normalized to be in the range 0 to +1 (or -1 to 1). For example rotation information on GameObjects should be recorded as `state.Add(transform.rotation.eulerAngles.y/180.0f-1.0f);` rather than `state.Add(transform.rotation.y);`. See the equation below for one approach of normaliztaion.
![normalization](../images/normalization.png)
## Actions
* When using continuous control, action values should be clipped to an appropriate range.

64
images/normalization.png

之前 之后
宽度: 697  |  高度: 104  |  大小: 14 KiB
正在加载...
取消
保存