mbunse / botui_demo

Chatbot Example

Backend

Installation

Python 3.8.10 is needed. Tested on Windows 10 64bit.

pip install -r requirements.txt
python -m spacy download en_core_web_md
python -m spacy download de_core_news_sm

Manual download of language packages:

curl -L --output de_core_news_md-3.0.0-py3-none-any.whl --url https://github.com/explosion/spacy-models/releases/download/de_core_news_md-3.0.0/de_core_news_md-3.0.0-py3-none-any.whl
curl -L --output en_core_web_md-3.0.0-py3-none-any.whl --url https://github.com/explosion/spacy-models/releases/download/en_core_web_md-3.0.0/en_core_web_md-3.0.0-py3-none-any.whl
pip install en_core_web_md-3.0.0-py3-none-any.whl
pip install de_core_news_md-3.0.0-py3-none-any.whl

Backend

The backend is found in the backend folder.

Train

Intent recognition and conversation flow prediction have to be trained

doit train

Run interactive

First, start the action server: doit run_action. Then:

doit interactive

Run Socket.io/REST server

doit run_actions
doit core_server

Frontend

Run the following command in the frontend folder

cd frontend
livereload -p 8080

The socketio frontend is available as http://localhost:8080/index.html, the REST version as http://localhost:8080/index_rest.html.

Date extraction

see https://rasa.com/docs/rasa/components#ducklingentityextractor

Deploy container in openshift

oc process -f deploy-openshift.yml | oc create -f -
oc start-build chatbot-core --from-dir . --follow

Connect to mongodb instance to check tracking

oc get pods
# find pod for mongodb and open terminal:
oc rsh chatbot-mongodb-1-23hsj
$ mongo 127.0.0.1:27017/$MONGODB_DATABASE -u $MONGODB_USER -p $MONGODB_PASSWORD
$ use sampledb
$ db.conversations.find()

Delete everything from OpenShift

oc delete all -l app=chatbot
oc delete all -l app=chatbot-mongodb
oc delete secret chatbot-mongodb
oc delete all -l build=chatbot-core
oc delete dc,route -l app=chatbot-core