search

Helsinki-NLP / OPUS-MT-train

Training open neural machine translation models
MIT License
318 stars 40 forks source link

en-es / es-en : spm instead of bpe? #16

Closed pentegroom closed 4 years ago

pentegroom commented 4 years ago

Hi,

Do you have spm versions of the tokenization for es-en / en-es models since source and target spm are required to convert to models into pytorch?

Thank you.

jorgtied commented 4 years ago

Unfortunately, I don't have any sentencepiece model at this moment but more models will come and soon be released. Most of them will appear from this repository: https://github.com/Helsinki-NLP/Tatoeba-Challenge

pentegroom commented 4 years ago

thank you so much.

jorgtied commented 4 years ago

There is now https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-spa and https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-eng

pentegroom commented 4 years ago

Thank you so much.

On Tue, Aug 18, 2020, 5:29 AM tiedemann notifications@github.com wrote:

There is now https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/eng-spa and https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/spa-eng

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/Helsinki-NLP/OPUS-MT-train/issues/16#issuecomment-675398692, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOQU2DTAQMTYTIPFP4GHAPLSBJJYBANCNFSM4PXKZSAA .