harvardnlp / seq2seq-attn

Sequence-to-sequence model with LSTM encoder/decoders and attention
http://nlp.seas.harvard.edu/code
MIT License
1.26k stars 278 forks source link

Blank the defaults for -targfile and fail if no -srcfile #13

Closed srush closed 8 years ago