suriyadeepan / practical_seq2seq

A simple, minimal wrapper for tensorflow's seq2seq module, for experimenting with datasets rapidly
http://suriyadeepan.github.io/2016-12-31-practical-seq2seq/
GNU General Public License v3.0
569 stars 270 forks source link

Predict with specific terms #4

Closed jonaslund closed 7 years ago

jonaslund commented 7 years ago

Hi

I was wondering if there's a straight forward way to the trained models to predict answers for specific terms / questions.

Since the model.predict takes encoded data, which is then decoded. Did you work out a method to encode sentences on the fly to get an interactive mode running?

Thx again

suriyadeepan commented 7 years ago

You can use some of the functions from this module. If you are interested in deploying the trained model to chat with users (without having to reinvent the wheel), you might wanna resuse some of the code from this repository. Specifically data_utils.py and chatbot.py.

jonaslund commented 7 years ago

Brilliant. Thanks