DeepQA Deployment for testing chatbots provisioned with Kubernetes - Work in progress original DeepQA source taken from:

• https://github.com/llSourcell/chatbot_tutorial baesd on:
• https://github.com/Conchylicultor/DeepQA

DISCLAIMER: !!! --- IMPORTANT --- !!! - The Dockerfile contains a secret key - you should NOT reuse this key or expose your key and manage secrets responsibly. This is a blind rehash with no prior knoweldge on most of the components used - use with caution and not in production.

This readme is a fast work in progress - basic - minimal

Usage: In kubernetes environemnt (this was build and tested in minikube) #Replace IP address for your local kubernetes deployment in the yml files then...

#to build out. ./run_once.sh

#to wipe/destroy ./wipe.sh

#post config. After iniital build if error about unable to find django session table then run migration on deepqa node. kubectl get pods (note deepqa pod name) kubectl exec -it -- /bin/bash python manage.py migrate

#To browse to bot example. http://192.168.99.101:8000/ #change ip to ip presented by your 'kubectl cluster-info' IP address.

#changes from branch hosted by Siraj (||Source||) - https://github.com/llSourcell/chatbot_tutorial

• created data/samples dir
• created missing save dir
• Downloaded missing movie_lines.txt and other files
• Installed pre-trained data
• updated to provision seperate redis server
• downgraded channels and django due to Conchylicultor/DeepQA#187 - recommended by wujiajia2017scnu

#TODO fix migrations so they don't need to be re-run after docker run make option for persistent redis storage/volumes. train box so it's not talking giberish.