Models
Cloudera AI Project Lifecycle
Models
Models - Concepts and Terminology
Challenges with Machine Learning in production
Challenges with model deployment and serving
Challenges with model monitoring
Challenges with model governance
Model visibility
Model explainability, interpretability, and reproducibility
Model governance using Apache Atlas
Using AI Registry
Setting up AI Registry
Creating a AI Registry
Creating a AI Registry on an Azure UDR Private Cluster
Setting up access for AI Registry in a RAZ-enabled environment
Setting up access for AI Registry in a non-RAZ-enabled environment
Synchronizing AI Registry with a workbench
Viewing details for AI Registries
AI Registry permissions
Model access control
Deleting AI Registry
Force delete a AI Registry
Registering and deploying models with AI Registry
Creating a model using MLflow
Registering a model using the AI Registry user interface
Registering a model using MLflow SDK
Using MLflow SDK to register customized models
Viewing registered model information
Creating a new version of a registered model
Deploying a model from the AI Registry page
Deploying a model from the AI Registry using APIv2
Deploying a model from the destination Project page
Viewing Details for AI Registry
Delete a model from AI Registry
Disabling AI Registry
Upgrade AI Registry
Roll back the registry upgrade
AI Registry standalone API
Prerequisites for AI Registry standalone API
Authenticating clients for interacting with AI Registry API
Role-based authorization
Using the REST Client
Importing external models from NGC Catalog to AI Registry
Importing external models from HuggingFace
AI Registry CLI Client
Known issues with AI Registry standalone API
Updating AI Registry configuration
Troubleshooting issues with AI Registry API
Cloudera AI Inference service cannot discover AI Registry
Debugging the model import failure
Creating and deploying a Model
Usage guidelines for deploying models with Cloudera AI
Known Issues and Limitations with Model Builds and Deployed Models
Request/Response Formats (JSON)
Testing calls to a Model
Securing Models
Access Keys for Models
API Key for Models
Enabling authentication
Generating an API key
Managing API Keys
Workflows for active Models
Technical metrics for Models
Debugging issues with Models
Deleting a Model
Managing Cloudera Copilot
Prerequisites to set up Cloudera Copilot
Configuring Cloudera Copilot
Using Cloudera Copilot
Example - Model training and deployment (Iris)
Training the Model
Deploying the Model