Rasa

Rasa is an open source machine learning framework. It helps you build contextual assistants that can have layered conversations with lots of back and forth. In order for a human to have a meaningful exchange with a contextual assistant, the assistant needs to be able to use context to build on things that have been previously discussed — Rasa enables you to build assistants that can do this in a scalable way.

With Rasa, you can build contextual assistants for the most popular messengers (Facebook, Slack and others), for your own conversational channels, and voice assistants like Alexa Skills and Google Home Actions.

You can deploy Rasa in your Nebius AI Managed Service for Kubernetes clusters using this Marketplace product.

Before installing this product:

1. Create a Kubernetes cluster and a node group in it. Each node in the group should have at least 8 GB of RAM.
2. install kubectl, and configure it to work with the created cluster.

To install the product:

1. On the cluster page in the management console, go to the Marketplace tab, select the product, and click Install.

2. Configure the application:

   • Namespace: Select a namespace or create a new one.
   • Application name: Enter an application name.
   • User: Create a user name to access the Weaviate UI.
   • Rasa token: Create and enter a token to authenticate the requests to the deployed model.
   • Model url: Add a link to the model archived in tar.gz format.
   • Replica count: Enter the required amount of replicas.
   • Enable Redis: Select to enable Redis database.
   • Enable Rabbit MQ: Select to enable Rabbit MQ event broker.
   • Enable Duckling: Select to enable Duckling entity extractor.
   • Enable Rasa action server: Select to enable the Action Server for deploying models.

3. Click Install.

4. Wait for the application to change its status to Deployed.

5. To check that Rasa is working:

   1. Run the commands below to start port forwarding:

      export RASA_SVC=$(kubectl get svc -l "app.kubernetes.io/name=rasa" -o name -n YOUR_NAMESPACE)
      kubectl --namespace YOUR_NAMESPACE port-forward $RASA_SVC 8000:80
      

   2. Send a test request:

      curl -XPOST http://127.0.0.1:8000/webhooks/rest/webhook \
      -H "Content-type: application/json" \
      -d "{
        \"message\": \"Hello\",
        \"metadata\": {\"token\": \"YOUR_RASA_TOKEN\"}
      }"
      

• Automating text- and speech-based conversations.

• Building contextual and voice assistants.

Nebius AI does not provide technical support for the product. If you have any issues, please refer to the developer’s information resources.