Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
ModelZoo
BERT_onnxruntime
Commits
c8f52cdd
Commit
c8f52cdd
authored
Oct 10, 2023
by
yangql
Browse files
Update README.md
parent
b8dc2f0f
Changes
1
Show whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
2 additions
and
0 deletions
+2
-0
README.md
README.md
+2
-0
No files found.
README.md
View file @
c8f52cdd
...
@@ -9,6 +9,7 @@
...
@@ -9,6 +9,7 @@
<p
align=
"center"
>
<p
align=
"center"
>
<img
src=
"./Doc/Images/Bert_01.png"
style=
"zoom:55%;"
align=
middle
>
<img
src=
"./Doc/Images/Bert_01.png"
style=
"zoom:55%;"
align=
middle
>
</p>
</p>
## 算法原理
## 算法原理
BERT的全称为Bidirectional Encoder Representation from Transformers,是一个预训练的语言表征模型。它强调了不再像以往一样采用传统的单向语言模型或者把两个单向语言模型进行浅层拼接的方法进行预训练,而是采用新的masked language model(MLM),以致能生成深度的双向语言表征。bert的输入可以是单一的一个句子或者是句子对,实际的输入值是segment embedding与position embedding相加,BERT的输入词向量是三个向量之和:Token Embedding:WordPiece tokenization subword词向量。
BERT的全称为Bidirectional Encoder Representation from Transformers,是一个预训练的语言表征模型。它强调了不再像以往一样采用传统的单向语言模型或者把两个单向语言模型进行浅层拼接的方法进行预训练,而是采用新的masked language model(MLM),以致能生成深度的双向语言表征。bert的输入可以是单一的一个句子或者是句子对,实际的输入值是segment embedding与position embedding相加,BERT的输入词向量是三个向量之和:Token Embedding:WordPiece tokenization subword词向量。
Segment Embedding:表明这个词属于哪个句子(NSP需要两个句子)。Position Embedding:学习出来的embedding向量。
Segment Embedding:表明这个词属于哪个句子(NSP需要两个句子)。Position Embedding:学习出来的embedding向量。
...
@@ -16,6 +17,7 @@ Segment Embedding:表明这个词属于哪个句子(NSP需要两个句子)
...
@@ -16,6 +17,7 @@ Segment Embedding:表明这个词属于哪个句子(NSP需要两个句子)
<p
align=
"center"
>
<p
align=
"center"
>
<img
src=
"./Doc/Images/Bert_04.png"
style=
"zoom:55%;"
align=
middle
>
<img
src=
"./Doc/Images/Bert_04.png"
style=
"zoom:55%;"
align=
middle
>
</p>
</p>
## 环境配置
## 环境配置
### Docker
### Docker
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment