Skip to content

Model Service

Overview

Model Service on Highrise Cloud allows users to deploy machine learning models as API services. This service enables easy deployment, API access, auto-scaling, and monitoring of model performance.

Deploying a Model

Step 1: Access Model Service

Log in to the Highrise Cloud platform and navigate to the Model Service section.

Step 2: Create a New Model Service

Click on the "New Model Service" button to start the deployment process.

Step 3: Configure Model Service

  • Name: Enter a name for your model service.
  • Model: Choose between an official or private model. Select the model from the dropdown menu.
  • Workers: Specify the number of workers needed for your service.

Tip

Choose the number of workers based on the expected load and performance requirements.

Step 4: Select Resources

Choose the appropriate resources for your model service. Consider the following options:

Attribute Description
CPU Type and number of CPUs
Memory Amount of memory
GPU Type and number of GPUs

Note

Consider the estimated cost and resource requirements carefully to optimize performance and cost.

Step 5: Confirm Deployment

Review your selections and click "Confirm" to deploy the model service.

Managing Model Services

Viewing Model Services

You can view all your deployed model services in the Model Services list. Here you can see the service status, resource usage, and creation time.

Deleting a Model Service

To delete a model service, click the "Delete" button next to the service you wish to remove.

Warning

Deletion is permanent and cannot be undone. Ensure that you no longer need the service before deleting.

Updating a Model Service

If you need to update the configuration of a model service, click the "Detail" button to access the service settings and make necessary changes.

Next Steps

After deploying your model service, you may want to explore Training Services to further develop and train your models, enhancing their performance and capabilities.

We wish you a successful experience with Model Service on Highrise Cloud!