The purpose of this repo is to showcase a contextual AI assistant built with the open source Rasa framework.
It also includes a docker-compose file for easy dev deployment.
To install, please clone the repo and run:
cd smalltalk-demo
pip install -r requirements.txt
pip install -e .
This will install the bot and all of its requirements. Note that this bot should be used with python 3.6 or 3.7.
Use rasa train
to train a model (this will take a significant amount of memory to train,
if you want to train it faster, try the training command with
--augmentation 0
).
Then, to run the rasa server, action server and duckling:
docker-compose up
- Integrate UI into docker-compose file
- Add tests