Out of the box integration with Rasa
Hybrid Chat provides support for the Rasa Open Source chatbot framework and Rasa X out of the box. Using our Rasa integration, you can create chatbots that are capable of answering queries, do back-office integrations to fetch customer data on the fly, create templates to support buttons and carousels on Facebook and Viber.
Hybrid Chat now supports bots trained with Rasa X which is a toolset provided by Rasa that layers on top of Rasa Open Source with a very good UI to help you train a chatbot.
Export Models and Actions to Rasa
- Go to the directory on your Google Drive (instructions provide in the notebook Part 1, Mounting Google drive ) where the models and actions are stored.
- Download the folders, 'actions' and 'models'
Upload these folders to the server where rasa-hybrid-bot is deployed. The path where you need to place these folders is,
<path to where chat-solution project is deployed>/chat-solution/docker/data/rasa-hybrid-bot/
- If these folders already exist on the server and you want to deploy a newly trained model, then replace them.
Go to <path to where chat-solution project is deployed>/chat-solution/docker/data/rasa-hybrid-bot/ and run the command
to change the permissions of the newly added folders.
Now you have to restart the rasa-hybrid-bot. Go to <path to where chat-solution project is deployed>/chat-solution and run the command
docker-compose -f docker/docker-compose-rasa-bot.yml restart rasa-hybrid-bot
Support for Custom and Standard Rasa 1.x
The Expertflow Hybrid Chat solution supports both Custom Rasa 1.x and Standard Rasa 1.x. Details regarding Rasa bot training, testing, and deployment: