Manage customized models
After creating a customized model through fine-tuning, reinforcement fine-tuning, distillation, or import, you can manage the model throughout its lifecycle. This section covers security configuration, monitoring, deployment, sharing across accounts, and troubleshooting.
Security and monitoring
Model customization access and security - Configure IAM roles, permissions, and encryption for model customization jobs.
Monitor your model customization job - Track the progress and status of your model customization jobs.
Analyze the results of a model customization job - Review metrics and results after a customization job completes.
Stop a model customization job - Stop a running model customization job if needed.
Using and deploying models
View details about a custom model - View details, metadata, and configuration of your custom models.
Set up inference for a custom model - Configure and deploy a custom model for inference.
Sharing and distribution
Share a model for another account to use - Share a custom model with another AWS account in your organization.
Copy a customized or shared model to use in a Region - Copy a customized or shared model to use in a different Region.
Maintenance
Delete a custom model - Delete a custom model when it's no longer needed.
Code samples for model customization - Example code for common model customization tasks.
Troubleshooting model customization issues - Diagnose and resolve common issues with model customization.
