Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
diluosixu
bert
提交
67a4537b
B
bert
项目概览
diluosixu
/
bert
与 Fork 源项目一致
从无法访问的项目Fork
通知
4
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
B
bert
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
未验证
提交
67a4537b
编写于
7月 15, 2019
作者:
S
Slav Petrov
提交者:
GitHub
7月 15, 2019
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
Update multilingual.md
Correct Wikipedia size correlation comment.
上级
0fce551b
变更
1
隐藏空白更改
内联
并排
Showing
1 changed file
with
4 addition
and
6 deletion
+4
-6
multilingual.md
multilingual.md
+4
-6
未找到文件。
multilingual.md
浏览文件 @
67a4537b
...
@@ -69,7 +69,7 @@ Note that the English result is worse than the 84.2 MultiNLI baseline because
...
@@ -69,7 +69,7 @@ Note that the English result is worse than the 84.2 MultiNLI baseline because
this training used Multilingual BERT rather than English-only BERT. This implies
this training used Multilingual BERT rather than English-only BERT. This implies
that for high-resource languages, the Multilingual model is somewhat worse than
that for high-resource languages, the Multilingual model is somewhat worse than
a single-language model. However, it is not feasible for us to train and
a single-language model. However, it is not feasible for us to train and
maintain dozens of single-language model. Therefore, if your goal is to maximize
maintain dozens of single-language model
s
. Therefore, if your goal is to maximize
performance with a language other than English or Chinese, you might find it
performance with a language other than English or Chinese, you might find it
beneficial to run pre-training for additional steps starting from our
beneficial to run pre-training for additional steps starting from our
Multilingual model on data from your language of interest.
Multilingual model on data from your language of interest.
...
@@ -152,11 +152,9 @@ taken as the training data for each language
...
@@ -152,11 +152,9 @@ taken as the training data for each language
However, the size of the Wikipedia for a given language varies greatly, and
However, the size of the Wikipedia for a given language varies greatly, and
therefore low-resource languages may be "under-represented" in terms of the
therefore low-resource languages may be "under-represented" in terms of the
neural network model (under the assumption that languages are "competing" for
neural network model (under the assumption that languages are "competing" for
limited model capacity to some extent).
limited model capacity to some extent). At the same time, we also don't want
to overfit the model by performing thousands of epochs over a tiny Wikipedia
However, the size of a Wikipedia also correlates with the number of speakers of
for a particular language.
a language, and we also don't want to overfit the model by performing thousands
of epochs over a tiny Wikipedia for a particular language.
To balance these two factors, we performed exponentially smoothed weighting of
To balance these two factors, we performed exponentially smoothed weighting of
the data during pre-training data creation (and WordPiece vocab creation). In
the data during pre-training data creation (and WordPiece vocab creation). In
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录