rasa-core-and-nlu-docker-compose

Rasa core and rasa nlu enviroment for local enviroment with moodbot example from official Rasa Stack documentation

Setup for Rasa Stack on local enviroment

Clone project:

git clone https://github.com/sasastarcevic/rasa-core-and-nlu-docker-compose.git

Go to project

cd rasa-core-and-nlu-docker-compose

Build docker-compose file

docker-compose build

Note: First time it can take 5-10 min!

Run docker-compose

docker-compose up -d

Check status of Rasa Stack container

docker ps -as

Rasa Stack container status

CONTAINER ID        IMAGE                COMMAND                  CREATED             STATUS              PORTS                                            NAMES                  SIZE
40ba123eecdb        rasacore_rasa_core   "./entrypoint.sh s..."   9 seconds ago       Up 8 seconds        0.0.0.0:5000->5000/tcp, 0.0.0.0:5005->5005/tcp   rasacore_rasa_core_1   525 kB (virtual 2.7 GB)

Attach to rasa_core container

docker exec -it 40ba123eecdb bash

Go to folder:

cd examples/moodbot/

Execute run-rasa-server.sh script and wait until Rasa Core server is started

./run-rasa-core.sh

Rasa Core server started sucessfully

Using TensorFlow backend.
2018-04-18 21:13:33+0000 [-] Log opened.
2018-04-18 21:13:33+0000 [-] Site starting on 5005
2018-04-18 21:13:33+0000 [-] Starting factory <twisted.web.server.Site object at 0x7fcca0c0abe0>

Check Rasa Nlu Rest API

curl 'http://localhost:5000/parse?q=hello'

Note: Official Rasa Nlu documentation

Check Rasa Core Rest API

curl -XPOST localhost:5005/conversations/default/parse -d '{"query":"hello there"}'

Note: Official Rasa Core documentation