Toggle navigation
Toggle navigation
此项目
正在载入...
Sign in
xuning
/
sherpaonnx
转到一个项目
Toggle navigation
项目
群组
代码片段
帮助
Toggle navigation pinning
Project
Activity
Repository
Pipelines
Graphs
Issues
0
Merge Requests
0
Wiki
Network
Create a new issue
Builds
Commits
Authored by
Fangjun Kuang
2023-09-20 07:49:53 +0800
Browse Files
Options
Browse Files
Download
Email Patches
Plain Diff
Committed by
GitHub
2023-09-20 07:49:53 +0800
Commit
6afa9c85f6a60549c822a12fafbd49d8c289b43a
6afa9c85
1 parent
bd173b27
Fix tokens for byte-level BPE token. (#324)
隐藏空白字符变更
内嵌
并排对比
正在显示
2 个修改的文件
包含
14 行增加
和
1 行删除
CMakeLists.txt
sherpa-onnx/csrc/symbol-table.cc
CMakeLists.txt
查看文件 @
6afa9c8
cmake_minimum_required
(
VERSION 3.13 FATAL_ERROR
)
project
(
sherpa-onnx
)
set
(
SHERPA_ONNX_VERSION
"1.7.1
5
"
)
set
(
SHERPA_ONNX_VERSION
"1.7.1
6
"
)
# Disable warning about
#
...
...
sherpa-onnx/csrc/symbol-table.cc
查看文件 @
6afa9c8
...
...
@@ -46,6 +46,19 @@ void SymbolTable::Init(std::istream &is) {
}
}
// for byte-level BPE
// id 0 is blank, id 1 is sos/eos, id 2 is unk
if
(
id
>=
3
&&
id
<=
258
&&
sym
.
size
()
==
6
&&
sym
[
0
]
==
'<'
&&
sym
[
1
]
==
'0'
&&
sym
[
2
]
==
'x'
&&
sym
[
5
]
==
'>'
)
{
std
::
ostringstream
os
;
os
<<
std
::
hex
<<
(
id
-
3
);
if
(
std
::
string
(
sym
.
data
()
+
3
,
sym
.
data
()
+
5
)
==
os
.
str
())
{
uint8_t
i
=
id
-
3
;
sym
=
std
::
string
(
&
i
,
&
i
+
1
);
}
}
assert
(
!
sym
.
empty
());
assert
(
sym2id_
.
count
(
sym
)
==
0
);
assert
(
id2sym_
.
count
(
id
)
==
0
);
...
...
请
注册
或
登录
后发表评论