forked from nosuggest/Reflection_Summary
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
shataowei
committed
Nov 25, 2019
1 parent
551ef88
commit 928fa5a
Showing
3 changed files
with
59 additions
and
41 deletions.
There are no files selected for viewing
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,20 @@ | ||
# 阐述CRF原理? | ||
- 首先X,Y是随机变量,P(Y/X)是给定X条件下Y的条件概率分布 | ||
- 如果Y满足马尔可夫满足马尔科夫性,及不相邻则条件独立 | ||
- 则条件概率分布P(Y|X)为条件随机场CRF | ||
|
||
# 线性链条件随机场的公式是? | ||
- ![](https://tva1.sinaimg.cn/large/006y8mN6gy1g9ah66y1nxj30cr015747.jpg) | ||
|
||
# CRF与HMM区别? | ||
- CRF是判别模型求的是p(Y/X),HMM是生成模型求的是P(X,Y) | ||
- CRF是无向图,HMM是有向图 | ||
- CRF全局最优输出节点的条件概率,HMM对转移概率和表现概率直接建模,统计共现概率 | ||
|
||
# Bert+crf中的各部分作用详解? | ||
- Bert把中文文本进行了embedding,得到每个字的表征向量 | ||
- dense操作得到了每个文本文本对应的未归一化的tag概率 | ||
- CRF在选择每个词的tag的过程其实就是一个最优Tag路径的选择过程 | ||
- CRF层能从训练数据中获得约束性的规则 | ||
- 比如开始都是以xxx-B,中间都是以xxx-I,结尾都是以xxx-E | ||
- 比如在只有label1-I,label2-I..的情况下,不会出现label1-B |