Ubuntu dependencies

sudo apt install -y python3-dev python3-pip python3-venv

Create python environment

Navigate to Rasa-bot
It is important to use environment as the dependencies for rasa conflict with the dependencies for speech IO

python3 -m venv ./rasa
source ./rasa/bin/activate

Pip Dependencies

pip3 install --upgrade pip
pip3 install -r requirements.txt

Run Rasa

Use rasa train to train model
Run rasa run actions --actions actions.actions first
Use rasa shell to run rasa in the terminal or use rasa run to run as a web service (use curl to test. See below)

Test Rasa REST API

curl -i -X POST -H "Content-Type: application/json" -d "{\"sender\":\"test_user\", \"message\":\"Hello Tiago\"}" http://localhost:5005/webhooks/rest/webhook

curl -i -X POST -H "Content-Type: application/json" -d "{\"sender\":\"test_user\", \"message\":\"Can you bring me the crackerbox\"}" http://localhost:5005/webhooks/rest/webhook

curl -i -X POST -H "Content-Type: application/json" -d "{\"sender\":\"test_user\", \"message\":\"Can you find my glasses?\"}" http://localhost:5005/webhooks/rest/webhook

Rasa Commands

rasa init

Creates a new project with example training data, actions, and config files.

rasa train Trains a model using your NLU data and stories, saves trained model in ./models.

rasa train --finetune Used for incremental training. Reuses previously trained model and retrains it using newly added data. It might still be useful to run a full training.

rasa run

rasa run --enable-api

Starts a server with your trained model.

rasa visualize

Generates a visual representation of your stories.

rasa interactive

Starts an interactive learning session to create new training data by chatting to your assistant.

rasa interactive --model ----skip-visualization Starts an interactive learning session without retraining model and generating visualizationras.

rasa shell or rasa shell --debug

Loads your trained model and lets you talk to your assistant on the command line.

rasa run actions --actions actions.actions

Run action server

docker run -p 8000:8000 rasa/duckling