practical-pytorch icon indicating copy to clipboard operation
practical-pytorch copied to clipboard

Published 20 hours ago • verified publisher icon spro

Can seq2seq model support the new sentence tranlslation?

Open quoniammm opened this issue 7 years ago • 5 comments

Why does the KeyError occurs?I can not understand. Can you help me?Thank you

quoniammm avatar Sep 11 '17 22:09 quoniammm

I guess you have not segment sentence, it's necessary for chinese.

czs0x55aa avatar Sep 12 '17 03:09 czs0x55aa

@czs0x55aa thanks!The problem is soved.I used Jieba to segment sentence.Do you have some tools or suggestions for Chinese tanslation(processing)?

quoniammm avatar Sep 12 '17 05:09 quoniammm

I'm think Jieba is the simplest tool for text segmentation, and work fine in most case. Maybe you would be interested in other tools, like NLPIR, THULAC-Python, ltp

czs0x55aa avatar Sep 12 '17 07:09 czs0x55aa

May I know how to support a new sentence translation? Jieba didn't solve my issue. I have run the model for English-Bengali translation.

paul-pias avatar Mar 11 '20 07:03 paul-pias

@paul-pias Inserting a space between each character in the sentences may help. For example: こんばんは -> こ ん ば ん は

Tad-Done avatar Nov 02 '20 08:11 Tad-Done