You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Great code!
I'm wondering which steps are necessary to move the greedy decoder from currently implemented to an actual beam search decoder.
Could you tell me which is the right point in the code where to add this functionality?
Hi,
Great code!
I'm wondering which steps are necessary to move the greedy decoder from currently implemented to an actual beam search decoder.
Could you tell me which is the right point in the code where to add this functionality?
refer http://stackoverflow.com/questions/37027665/implementing-seq2seq-with-beam-search
and tensorflow/tensorflow#654
https://github.com/tensorflow/tensorflow/blob/34df7ee6f5a8f931d2433dc7e6e739bc880a35ea/tensorflow/contrib/seq2seq/python/ops/decoder.py the dynamic_decode was built to ,with beam search
Thanks a lot!
The text was updated successfully, but these errors were encountered: