Skip to content

spongepad/Korean-Honorific-Translation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Korean Honorific Neural Machine Translation

The repo is a model that converts Korean into polite sentences.(using OpenNMT Transformer)

Getting Started

Step 1. Preprocess

!python preprocess.py

The source text file(src) and target text file(tgt)
default tokenize : Mecab+SentencePiece.

Step 2. Train

!python train.py

If you want to continue training the model, add --train_from (model path)/model.pt later.

Step 3. Translate

!python translate.py -model data/model/model.pt -src data/src-test.txt -tgt data/tgt-test.txt -replace_unk -verbose -gpu 0

Step 4. Postprocess

!python ./onmt/tools/spacing.py -i ./data/pred.txt -o ./data/pred_sp.txt

Step 5. Scoring the model

!perl tools/multi-bleu.perl data/tgt-test.txt < data/pred.txt

Reference

https://github.com/OpenNMT/OpenNMT-py

Releases

No releases published

Packages

No packages published