Vijay21
(Vijay)
September 16, 2019, 7:25am
1
I have trained my NLU model using the tensorflow model. When i try to predict the model.
it keep on loading without timeout… it got stuck on the parse method of rasa nlu python library.
This issue is happening when i am running this as docker container. I am not getting any error to debug this issue. Since the request is not getting timedout.
I am using the below versions
python : 3.7.3
rasa_nlu: 0.14.6
tensorflow: 1.13.1
tyd
(Ty Dunn)
September 16, 2019, 7:08pm
2
Hi @Vijay21 ! Have you tried testing the model on your local machine using rasa shell nlu ?
Vijay21
(Vijay)
September 17, 2019, 2:15am
3
Hi @tyd Yes I have tested my model using rasa nlu shell in my local machine
tyd
(Ty Dunn)
September 17, 2019, 11:52am
4
@Vijay21 And that model works when you tested it? You are only having issues when deployed on Docker?
Vijay21
(Vijay)
September 19, 2019, 6:34am
5
Yes I have issue when I am running it in docker in my linux server @tyd
tyd
(Ty Dunn)
September 19, 2019, 7:44am
6
@Vijay21 have you tried upgrading to a newer version of Rasa NLU?
Vijay21
(Vijay)
September 23, 2019, 5:52am
7
@tyd no I have not Trier upgrading to newer Version. . I have tried with docker Instance given by rasa.
msamogh
(Amogh Mannekote)
September 24, 2019, 7:15pm
8
@Vijay21 I strongly recommend you upgrade to the latest version of Rasa and give it a try and let us know how it goes.