Deploying a model

How to deploy a model

To use your model with your application, you need to deploy it.

To deploy a model:

Navigate to NLU → Deployments section and click Deploy.
In the opened dialog box, select a deployment type, either Testing or Production, and click Next.

Never use Testing deployment for production! They should only be used for testing purposes.

In the opened dialog box, fill in the fields as described below.

Name: Name for your application.
Group: A group of users who can use the deployed model.
Description: An optional description for the deployment.

In the next dialog box, select a model from the list of available models. Alternatively, you can import a previously exported model. To import a model, drop a ZIP file directly or click Browse Files.

5. Click Deploy. The deployment is added to the Deployments list. While the deployment process is ongoing, its status is set as Working.

6. As soon as the process is completed, the status will automatically change to Running. You can switch on the Created by me toggle, to filter out the deployments created by you.

Testing deployments are automatically stopped after 30 days. Before the deployment is stopped, you will receive automated service notifications 24 hours and 2 hours in advance.

You can start a stopped deployment manually at any time.

The completed Testing deployments are marked with the Bug icon.

The imported models are marked with the Upload icon.

If you navigate to the NLU Models tab, you will also see the deployed models labeled with the Rocket icon for your convenience.

Model deployments details

To see the model’s deployment details, proceed as follows:

Navigate to NLU → NLU Models.
Select a model from the list and click on it. The drill-down page opens.
Go to the Deployments tab. All the model deployments are listed here.
Click the Arrow button to navigate to the deployment management page and see more information on a specific deployment.