No Description
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Andreas Peters dd97a2de3d Merge branch 'v2.2.1.av1' of AVENTER/nlp-backend into master 6 months ago
app Fix return type to prevent promise error (#98) 7 months ago
examples updated restaurent search data 10 months ago
frontend Few other minor changes on the front-end for entities (#99) 7 months ago
logs updated gitignore 11 months ago
model_files finsihed starspace integration 10 months ago
tests Add travis config file (#50) 1 year ago
.dockerignore updated gitignore 11 months ago
.gitignore updated gitignore 11 months ago
.travis.yml Add travis config file (#50) 1 year ago
License.txt Added MIT License 1 year ago
Makefile some more flexible 6 months ago
Procfile Create Procfile 1 year ago Update 10 months ago
app.json updated app.json to add heroku build packs 1 year ago some more flexible 6 months ago
docker-compose.yml added slider for intent detection threshold 11 months ago
dockerfile some more flexible 6 months ago fix some typo and docstring (#88) 8 months ago
requirements.txt finished tensorflow intent classifier. need to clean and fine tune 10 months ago added dropouts to reduce overfitting 10 months ago


Build Status

An AI Chatbot framework built in Python

Building a chatbot can sound daunting, but it’s totally doable. IKY is an AI powered conversational dialog interface built in Python. With IKY it’s easy to create Natural Language conversational scenarios with no coding efforts whatsoever. The smooth UI makes it effortless to create and train conversations to the bot and it continuously gets smarter as it learns from conversations it has with people. IKY can live on any channel of your choice (such as Messenger, Slack etc.) by integrating it’s API with that platform.

You don’t need to be an expert at artificial intelligence to create an awesome chatbot that has artificial intelligence. With this basic project you can create an artificial intelligence powered chatting machine in no time.There may be scores of bugs. So feel free to contribute via pull requests.


Using docker-compose (Recommended)

docker-compose build
docker-compose up -d
docker-compose exec iky_backend python init

Using Docker

# build docker images
docker build -t iky_backend:2.0.0 .
docker build -t iky_gateway:2.0.0 frontend/.

# start iky backend
docker run --name=iky_backend -e="APPLICATION_ENV=Production" iky_backend:2.0.0

# setup default intents
docker exec -it iky_backend python init

# start iky gateway with frontend
docker run --name=iky_gateway --link iky_backend:iky_backend -p 8080:80 iky_gateway:2.0.0

without docker


  • Setup Virtualenv and install python requirements ```sh make setup

make run_dev

source venv/bin/activate && python init

* Production
make run_prod


  • Development sh cd frontend npm install ng serve
  • Production sh cd frontend ng build --prod --environment=python serve files in dist/ folder using nginx or any webserver



  • add your dev/production configurations in



You can import some default intents using follwing steps



Checkout this basic tutorial on youtube,


Watch tutorial on Fullfilling your Chatbot Intent with an API Call - Recipe Search Bot

Please visit my website to see my personal chatbot in action


  • Write Unit Tests
  • PEP-8 compliance
  • Word2Vec Integration
  • NLTK to Spacy migration
  • PyCRFSuite to sklearn-crfsuite migration
  • Support follow up conversations

### Dependencies documentations

Free Software, Hell Yeah!

Made with :heart: at God’s Own Country.