bowenwen/rasa_chatbot_nb repository overview

⁠Transit Chatbot

A friendly trip planning chatbot that provides transit directions from A to B and information on the next bus arrival time.

Developed using the open-source conversational AI library Rasa stack⁠. The framework is generic, however, the implementation is specific to transit services in Metro Vancouver, BC, Canada offerred by TransLink.

The chatbot is not an official TransLink product.

⁠Getting started: test chatbot in Jupyter with docker

⁠Get a copy on your local

git clone https://github.com/moh-salah/transit_chatbot.git

⁠Start container

docker-compose up

You can access the Jupyter Notebook working environment on: http://localhost:8888/.

You can check the status of the rasa action server on: http://localhost:5055/.

You can test the pre-trained chatbot in the terminal within the docker container environment: python -m rasa_core.run -d models/dialogue -u models/nlu/default/current --endpoints endpoints.yml --debug

⁠Cleanup containers

docker-compose down

⁠Deploying the chatbot with docker

Deploying trained chatbot is easy with rasa prebuilt images

⁠Get a copy on your local

git clone https://github.com/moh-salah/transit_chatbot.git

cd rasa_docker

⁠Retrain the model with Rasa Docker (Optional)

⁠Train rasa core

docker run -v $pwd/:/app/project -v $pwd/models/rasa_core:/app/models rasa/rasa_core:0.12.3 train --domain project/domain.yml --stories project/data/stories.md --config project/policy_config.yml --out models

⁠Train rasa nlu

docker run -v $pwd/:/app/project -v $pwd/models/rasa_nlu:/app/models rasa/rasa_nlu:0.13.8-spacy run python -m rasa_nlu.train -c project/config.yml -d project/data/data.json -o models --project current

⁠Build a custome Rasa action image (Optional)

docker build --tag my_action_image .

If you decide to build your own image, please modify the docker-compose file accordingly.

⁠Start chatbot services

docker-compose up -d

You need to set up your own chatbot platform connectors with relevant credentials, read more about connectors here⁠.

While you are configuring your connectors, you might want to use a tunnel service if you are not configuring on a server or cloud service with public domain. ngnork⁠ offers free tunnel service.

⁠Cleanup containers

docker-compose down

For more information on working with Rasa Docker Images, check out their guide⁠.