Great! How's the results?

[hunkim]

I was wondering which one works the best. Thanks in advance.

First trial: seq2seq w/ attention + nucle 2.3 dataset
Next..
more data, data augmentations, ...
tweeking (tweaking) nets like Bytenet or Transformer 
beam search

[andabi]

@hunkim The results was in fact not so successful. even though we have other ideas in mind and more datasets, we didn't try those yet haha..

[hunkim]

Thanks!

[andabi]

@hunkim when i make a progress, i'll let you know. thanks for your interest ;)

[andabi]

@hunkim i made a mistake. The results was not so successful.