zhongkaifu
commited on
Commit
•
3d2041f
1
Parent(s):
a647159
Update README.md
Browse files
README.md
CHANGED
@@ -7,4 +7,16 @@ tags:
|
|
7 |
- translation
|
8 |
---
|
9 |
This is a translation model from English to Chinese and it can be consumed by Seq2SeqSharp (https://github.com/zhongkaifu/Seq2SeqSharp).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
|
|
|
7 |
- translation
|
8 |
---
|
9 |
This is a translation model from English to Chinese and it can be consumed by Seq2SeqSharp (https://github.com/zhongkaifu/Seq2SeqSharp).
|
10 |
+
You can clone Seq2SeqSharp, build it and run the model as follows:
|
11 |
+
1. git clone https://github.com/zhongkaifu/Seq2SeqSharp.git
|
12 |
+
2. cd Seq2SeqSharp
|
13 |
+
3. dotnet build Seq2SeqSharp.sln --configuration Release
|
14 |
+
4. cd Tools
|
15 |
+
5. cd Seq2SeqConsole
|
16 |
+
6. cd bin
|
17 |
+
7. ./bin/Seq2SeqConsole -Task Test -ModelFilePath mt_enu_chs.model -InputTestFile %InputFilePath% -OutputFile %OutputFilePath% -ProcessorType %Device% -MaxSrcSentLength 220 -MaxTgtSentLength 220
|
18 |
+
8. %InputFilePath% is English input file for translation. It's one sentence per line. It needs to be tokenized by SentencePiece encoder using enuSpm.model in this repo
|
19 |
+
9. %OutputFilePath% is translated Chinese file output by the model. It's one sentence per line. You can decode them by SentencePiece decoder using cjkSpm.model in this repo
|
20 |
+
10. %Device% is device type for model inference, it can be GPU, CPU and CPU_MKL(For Windows only).
|
21 |
+
11.
|
22 |
|