search

stanfordnlp / cocoa

Framework for learning dialogue agents in a two-player game setting.
MIT License
158 stars 62 forks source link

make system compatiable with tensorflow1.1 #37

Closed zilongzheng closed 5 years ago

zilongzheng commented 7 years ago

modify model syntax to make training part work under tensorflow1.1.0 modify chat.html to change variable ctr to loop.index, variable "ctr" does not change over loop in my test