A practical and feature-rich paraphrasing framework to augment human intents in text form to build robust NLU models for conversational engines. Created by Prithiviraj Damodaran. Open to pull requests and other forms of collaboration.
Apache License 2.0
867
stars
143
forks
source link
Adding condition to skip adequacy and fluency models for thresholds <= 0 #8
Hello,
I added two simple if else statements to skip the adequacy and fluency filtering when the thresholds are <= 0. When they are deactivated with this addition the generation is around 3/4 times faster.
Thanks!
Hello, I added two simple if else statements to skip the adequacy and fluency filtering when the thresholds are <= 0. When they are deactivated with this addition the generation is around 3/4 times faster. Thanks!