Rasa NLU/Core Weather Bot
Current configuration supports training, command-line interface and HTTP server API.
- pip install rasa_nlu
- pip install rasa-core
- pip install weather-api
First you need to train your bot. You can do that by simply running: python trainer.py train-all
This will train both the NLU and Stories using Tensorflow machine learning.
To run in command line just call: python bot.py run
To run as a server you need to call: python -m rasa_core.server -d models/dialogue -u models/nlu/default/current -o models/out.log
You talk to the server through API as: curl -XPOST localhost:5005/conversations/default/respond -d '{"query":"hi"}'
Documentation: http://rasa.com/docs/core/http/
- Rasa NLU: http://rasa.com/docs/nlu/
- Rasa Core: http://rasa.com/docs/core/
- Weather API: https://pypi.org/project/weather-api/ (Yahoo! Weather)