Homepage/Cloudera AI1.5.4

(latest)

Search Documentation

▶︎Cloudera
1. Reference Architectures
▶︎Cloudera on cloud
▼Cloudera on premises
▶︎Cloudera Manager
1. Cloudera Manager
▶︎Applications
▶︎Legacy
▶︎

Data Services
Getting Started
Cloudera Manager
Management Console
Replication Manager
Data Catalog
Data Engineering
Data Warehouse
Data Warehouse Runtime
Cloudera AI
Base
Getting Started
Runtime & Cloudera Manager
Upgrade
Storage
Flow Management
Streaming Analytics
CFM Operator
CSA Operator
CSM Operator

«

Filter topics

▶︎All roles
▼Data Scientist
▶︎Administrator
1. ▶︎Managing your Cloudera AI Workbench service
2. ▶︎Supporting your Cloudera AI service
  1. ▶︎Troubleshooting for the Administrator

(Optional) Configure the Sync between Cloudera AI and PyCharm
(Optional) Using VS Code with Git integration
(Optional) Using VS Code with Jupyter
(Optional) Using VS Code with Python
(Optional) Using VS Code with R
About Cloudera AI
Access Keys for Models
Access Keys for Models
Access the Terminal
Accessing data with Spark
Accessing Environmental Variables from projects
Accessing environmental variables from projects
Accessing local files in Ozone
Accessing Ozone storage
Add a catalog
Add Cloudera AI as an Interpreter for PyCharm
Add Docker registry credentials
Adding a Collaborator
Adding an SSH key to GitHub
Adding Custom ML Runtimes through the Runtime Catalog
Adding Docker registry credentials and certificates
Adding Hadoop CLI to ML Runtime sessions
Adding ML Runtimes using Runtime Repo files
Adding new ML Runtimes
Adding Project Collaborators
Adding Spark to ML Runtime Sessions
Adding the new ML Runtime
Administrator
All roles
AMP Project Specification
AMPs in airgapped environments
Analytical Applications
Analytical Applications
Apache Spark supported versions
API
API Key for Models
API Key for Models
API permissions for Projects
API v2 usage
Application polling endpoint
Applications
Applied ML Prototypes (AMP)
Authenticating Hadoop for Cloudera AI service accounts
Auto discovering data sources
Automatic Logging
Backing up a Cloudera AI Workbench
Backing up Cloudera AI Workbenches
Base Engine 10-cml1.3
Base Engine 11-cml1.4
Base Engine 12-cml-2020.06-2
Base Engine 13-cml-2020.08-1
Base Engine 14-cml-2021.05-1
Base Engine 15-cml-2021.09-1
Base Engine 9-cml1.2
Build and deployment commands for models
Build the new Docker image
Building the new Docker Image
Built-in Cloudera AI Visualizations
Built-in Cloudera AI Visualizations
Burstable CPUs
Business Users and Cloudera AI
Business Users and Cloudera AI
Catalog File Specification
cdswctl Command Line Interface client
cdswctl command reference
Certified Scale Limitations for Cloudera AI Workbenches on Cloudera Embedded Container Service
Challenges with Machine Learning in production
Challenges with model deployment and serving
Challenges with model governance
Challenges with model monitoring
Choosing default engine
CLI
Cloudera Accelerators for Machine Learning Projects
Cloudera AI and outbound network access
Cloudera AI API v2
Cloudera AI email notifications
Cloudera AI Experiment Tracking through MLflow API
Cloudera AI Metrics Collector Service
Cloudera AI Metrics Collector Service overview
Cloudera AI on premises 1.5.4 CHF1
Cloudera AI on premises 1.5.4 CHF2
Cloudera AI on premises 1.5.4 CHF3
Cloudera AI on premises 1.5.4 SP1
Cloudera AI on premises 1.5.4 SP2
Cloudera AI on premises 1.5.4 SP2 CHF1
Cloudera AI Overview
Cloudera AI Project Lifecycle
Cloudera AI Quota Management dashboard
Cloudera AI Registry permissions
Cloudera AI Workbenches
Cloudera Data Visualization for Cloudera AI
Cloudera Data Visualization for Cloudera AI
Collaboration models
Collaboration Models
Command Line Tools in Cloudera AI
Configure a browser-based IDE as an Editor
Configure PyCharm as a local IDE
Configure PyCharm to use Cloudera AI as the remote console
Configure VS Code as a local IDE
Configuring a browser-based IDE at the Project level
Configuring a local IDE using an SSH gateway
Configuring GPU usage
Configuring HTTP Headers for Cloudera AI
Configuring Project-level Runtimes
Configuring Quotas
Configuring the engine environment
Connect to external SQL databases
Connecting to Cloudera Data Warehouse
Connecting to external Amazon S3 buckets
Connecting to Hive and Impala services on Cloudera on premises Base
Connecting to Hive tables via HWC
Connecting to Iceberg tables
Connecting to Ozone filesystem
Connecting to Ozone filesystem
Controlling User access to features
Create a Dockerfile for the custom image
Creating a Cloudera AI Registry
Creating a customized engine image
Creating a Dockerfile for the custom Runtime Image
Creating a Job
Creating a machine user and synchronizing to workbench
Creating a model
Creating a model using MLflow
Creating a model using the model CLI
Creating a new version of a registered model
Creating a Pipeline
Creating a Project
Creating a Project
Creating a project from a password-protected Git repo
Creating a resource pool for Cloudera AI
Creating a Team
Creating a Team
Creating an Ozone data connection
Creating an SSH tunnel
Creating and deploying a Model
Creating and deploying a Model
Creating customized ML Runtimes
Creating New AMPs
Creating Resource profiles
Creating Resource profiles
Cumulative Hotfixes: Cloudera AI on premises
Custom AMP Catalog
Custom Runtime addons with Cloudera AI
Customized engine images
Customized Runtimes
Customizing the editor
Data Access
Data connection management
Data Scientist
Debugging issues with Models
Delete a model from Cloudera AI Registry
Deleting a Model
Deleting a Project
Deleting Cloudera AI Registry
Deploying a Cloudera AI Workbench with Support for TLS
Deploying a model from the AI Registry page
Deploying a model from the Cloudera AI Registry using APIv2
Deploying a model from the destination Project page
Deploying a new model with updated resources
Deploying an MLflow model as a Cloudera AI Model REST API
Deploying the Cloudera AI Workbench model
Deploying the Model
Differences Between Cloudera on cloud and Cloudera on premises
Disable addons
Disable or deprecate Runtime addons
Disabling and Deleting Runtimes
Disabling Cloudera AI Registry
Distribute the image
Distributed Computing
Distributed Computing with Workers
Distributing the ML Runtime Image
Dockerfile compatible with PBJ Workbench
Documenting Your Analysis
Documenting Your Analysis
Download cdswctl and add an SSH Key
Downloading and configuring cdswctl
Downloading diagnostic bundles for a workbench
Editors
Embedded Web Applications
Enable Cross-Origin Resource Sharing (CORS)
Enable HTTP security headers
Enable HTTP Strict Transport Security (HSTS)
Enabling authentication
Enabling authentication
Enabling model governance
Enabling model metrics
Enabling Quota Management in Cloudera AI
End-to-end example: MeCab
Engine environment variables
Engine environment variables
Environmental Variables
Ephemeral storage
Example - Model training and deployment (Iris)
Example models with PBJ Runtimes
Example: A Shiny Application
Example: Flask application
Example: Locating and adding JARs to Spark 2 configuration
Example: Monte Carlo estimation
Experiments
Experiments with MLflow
Exploratory Data Science and Visualization
Exploratory Data Science and Visualization
Export Usage List
Fixed Issues
Fixed issues in 1.5.4 SP2 CHF1
Force delete a Cloudera AI Registry
Generating an API key
Generating an API key
Getting Started with Cloudera AI on premises
Git for Collaboration
GPU usage
GPUs
Grid Displays
Grid Displays
Hadoop authentication for Cloudera AI Workbenches
Handling project volume size increase in Cloudera AI
Heterogeneous GPU clusters
Host name required by Learning Hub
Host names required by AMPs
Hosting an LLM as a Cloudera AI Workbench model
How to make base cluster configuration changes
HTML Visualizations
HTML Visualizations
HuggingFace Spaces and Community AMPs
IFrame Visualizations
IFrame Visualizations
Including images in allowlist for Cloudera AI projects
Initialize an SSH connection to Cloudera AI for VS code
Initializing an SSH endpoint
Installing a Jupyter extension
Installing a Jupyter kernel
Installing additional ML Runtimes Packages
Installing additional packages
Internal Network File System on Cloudera Embedded Container Service
Internal Network File System on OpenShift Container Platform
Introduction to Cloudera on premises
Jobs and Pipelines
Jobs API
Jupyter Magic Commands
Jupyter Magics
Known Issues
Known issues and limitations
Known Issues and Limitations in ML Runtimes older releases
Known Issues and Limitations in ML Runtimes version 2024.05.01
Known Issues and Limitations in ML Runtimes version 2024.05.02
Known Issues and Limitations in ML Runtimes version 2024.10.01
Known Issues and Limitations with Model Builds and Deployed Models
Launch a Session
Legacy Engine level configuration
Legacy Engines
Limitations for Quota Management
Limitations on Cloudera on premises
Limitations to customized ML Runtime images
Limitations with Analytical Applications
Limitations with customized engines
Limiting files in Explorer view
Linking an existing Project to a Git remote
Listing ML Runtimes
Logging into cdswctl
Managing a Team Account
Managing a Team Account
Managing API Keys
Managing API Keys
Managing default and backup data connections
Managing dependencies for Spark 2 and Scala
Managing dependencies for Spark 2 jobs
Managing Engines
Managing memory available for Spark drivers
Managing ML Runtimes
Managing Models
Managing Project Files
Managing Projects
Managing Projects
Managing Users
Managing your Personal Account
Managing your Personal Account
Manual data connections to connect to data sources
Metadata for custom ML Runtimes
ML Runtimes 2020.11
ML Runtimes 2021.12
ML Runtimes 2022.04
ML Runtimes 2022.11
ML Runtimes 2023.08.2
ML Runtimes 2023.12.1
ML Runtimes 2024.02.1
ML Runtimes 2024.05.1
ML Runtimes 2024.05.2
ML Runtimes 2024.10.1
ML Runtimes environment variables
ML Runtimes Environment Variables List
ML Runtimes Known Issues and Limitations
ML Runtimes NVIDIA GPU Edition
ML Runtimes NVIDIA RAPIDS Edition
ML Runtimes Pre-installed Packages
ML Runtimes Pre-installed Packages overview
ML Runtimes Pre-installed Packages overview
ML Runtimes Release Notes
ML Runtimes Version 2020.11
ML Runtimes Version 2021.02
ML Runtimes Version 2021.04
ML Runtimes Version 2021.06
ML Runtimes Version 2021.09
ML Runtimes Version 2021.09.02
ML Runtimes Version 2021.12
ML Runtimes Version 2022.04
ML Runtimes Version 2022.11
ML Runtimes Version 2022.11.2
ML Runtimes Version 2023.05.1
ML Runtimes Version 2023.05.2
ML Runtimes Version 2023.08.1
ML Runtimes Version 2023.08.2
ML Runtimes Version 2023.12.1
ML Runtimes versus Legacy Engine
ML Runtimes What's New
MLflow transformers
Model access control
Model explainability, interpretability, and reproducibility
Model Governance
Model Governance Requirements
Model governance using Apache Atlas
Model Metrics
Model visibility
Models
Models - Concepts and Terminology
Modes of configuration
Modifying Project settings
Monitoring active Models
Monitoring and alerts
Monitoring applications
Monitoring Cloudera AI Workbenches
Monitoring User Activity
Monitoring User Events
Native Workbench Console and Editor
Network File System (NFS)
NFS Options for on premises
Non-user dependent Resource Usage Limiting for workloads
NTP proxy setup on Cloudera AI
Onboarding Business Users
PBJ R 3.6 Libraries
PBJ R 4.0 Libraries
PBJ R 4.1 Libraries
PBJ Runtimes and Models
PBJ Workbench
Personal key
Pre-Installed Packages in engines
Preparing to manage models for using the model CLI
Prerequisites for Cloudera AI Discovery and Exploration
Prerequisites for creating Cloudera AI Registry
Project garbage collection
Provisioning a Cloudera AI Workbench
Provisioning a workbench using a configured resource pool
Python
Python 3.10 Libraries for Conda
Python 3.10 Libraries for Conda
Python 3.10 Libraries for Conda
Python 3.10 Libraries for Conda
Python 3.10 Libraries for Conda
Python 3.10 Libraries for Conda
Python 3.10 Libraries for JupyterLab
Python 3.10 Libraries for JupyterLab
Python 3.10 Libraries for JupyterLab
Python 3.10 Libraries for JupyterLab
Python 3.10 Libraries for JupyterLab
Python 3.10 Libraries for JupyterLab
Python 3.10 Libraries for Workbench
Python 3.10 Libraries for Workbench
Python 3.10 Libraries for Workbench
Python 3.10 Libraries for Workbench
Python 3.10 Libraries for Workbench
Python 3.10 Libraries for Workbench
Python 3.11 Libraries for JupyterLab
Python 3.11 Libraries for JupyterLab
Python 3.11 Libraries for JupyterLab
Python 3.11 Libraries for JupyterLab
Python 3.11 Libraries for JupyterLab
Python 3.11 Libraries for Workbench
Python 3.11 Libraries for Workbench
Python 3.11 Libraries for Workbench
Python 3.11 Libraries for Workbench
Python 3.12 Libraries for JupyterLab
Python 3.12 Libraries for Workbench
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for JupyterLab
Python 3.7 Libraries for PBJ Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.7 Libraries for Workbench
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for JupyterLab
Python 3.8 Libraries for PBJ Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.8 Libraries for Workbench
Python 3.9 Libraries for JupyterLab
Python 3.9 Libraries for JupyterLab
Python 3.9 Libraries for JupyterLab
Python 3.9 Libraries for JupyterLab
Python 3.9 Libraries for JupyterLab
Python 3.9 Libraries for JupyterLab
Python 3.9 Libraries for PBJ Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9 Libraries for Workbench
Python 3.9.6 Libraries for JupyterLab
Python 3.9.6 Libraries for JupyterLab
Python 3.9.6 Libraries for JupyterLab
Querying the engine type
Quota for Cloudera AI workloads
Quota Management overview
Quota Management [Technical Preview]
R 3.6 Libraries
R 3.6 Libraries
R 3.6 Libraries
R 3.6 Libraries
R 4.0 Libraries
R 4.0 Libraries
R 4.0 Libraries
R 4.0 Libraries
R 4.1 Libraries
R 4.1 Libraries
R 4.1 Libraries
R 4.1 Libraries
R 4.3 Libraries
R 4.3 Libraries
R 4.3 Libraries
R 4.4 Libraries
R 4.4 Libraries
R 4.4 Libraries
Recommended troubleshooting workflow
Registering a model using MLflow SDK
Registering a model using the Cloudera AI Registry user interface
Registering and deploying models with Cloudera AI Registry
Registering training data lineage using a linking file
Release Notes
Removing Cloudera AI Workbenches
Replace a Certificate
Repository Locations for 1.5.4 SP2 CHF1
Request/Response Formats (JSON)
Requirements
Requirements for Cloudera AI on Cloudera Embedded Container Service
Requirements for Cloudera AI on OpenShift Container Platform
Resource Usage Dashboard
Restarting a failed AMP setup
Restore a Cloudera AI Workbench
Restrictions for upgrading R and Python packages
Run Code
Running an Experiment using MLflow
Running Spark with Yarn on the Cloudera base cluster
Running workloads using a service account
Runtimes
Saved Images
Saved Images
Scala
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Scala 2.11 Libraries for Workbench
Securing Applications
Securing Models
Securing Models
Securing Models
Security
Service Accounts
Service Packs: Cloudera AI on premises
Set up a custom repository location
Setting Permissions for an Experiment
Setting Resource Usage Limiting for workloads
Setting up a Hive or Impala data connection manually
Setting up a Spark data connection
Setting up an HTTP Proxy for Spark 2
Setting up Cloudera AI Registry
Setting up the GPU node
Setting up VS Code
Sharing Job and Session Console Outputs
Simple Plots
Simple Plots
Site Administration
Spark
Spark configuration files
Spark Log4j configuration
Spark on Cloudera AI
Spark web UIs
SSH Keys
Starting Data Discovery and Visualization
Starting sessions and creating SSH endpoints
Stop a Session
Synchronizing Cloudera AI Registry with a workbench
Synchronizing machine users from the Synced team
Team key
Technical metrics for Models
Test Your Connectivity to the Cloudera-Data Center Cluster
Testing a browser-based IDE in a Session
Testing calls to a Model
Testing GPU Setup
Testing ML Runtime GPU Setup
Third-party Editors
Top Tasks
Tracked User events
Tracking metrics for deployed models
Tracking model metrics without deploying a model
Training the Model
Troubleshooting custom Runtime addons
Troubleshooting for Cloudera AI Metrics Collector Service
Troubleshooting for ML Runtimes
Troubleshooting for the Administrator
Troubleshooting for the Data Scientist
Troubleshooting issues with workloads
Troubleshooting Kerberos issues
Troubleshooting: 401 Unauthorized
Troubleshooting: 401 Unauthorized when accessing Hive or Impala virtual warehouses
Troubleshooting: Empty data page
Troubleshooting: Existing connection name
Turning off ML Runtimes Addons
Updating ML Runtime images on Cloudera AI installations
Updating proxy configuration in an existing workbench
Upgrading Cloudera AI Workbenches
Uploading and working with local files
Usage guidelines for deploying models with Cloudera AI
User Roles
User Roles
User Roles and Team Accounts
Using an External NFS Server
Using an MLflow Model Artifact in a Model REST API
Using Cloudera AI Registry
Using Conda Runtime
Using Conda to manage dependencies
Using data connection snippets
Using Editors for ML Runtimes
Using GPUs for Cloudera AI projects
Using JDBC Connection with PySpark
Using JupyterLab with ML Runtimes
Using ML Runtimes addons
Using ML Runtimes with cdswctl
Using MLflow SDK to register customized models
Using Runtime Catalog
Using Spark 2 from Python
Using Spark 2 from Scala
Using Spark 3 from R
Viewing details for Cloudera AI Registries
Viewing Details for Cloudera AI Registry
Viewing Job History
Viewing lineage for a model deployment in Atlas
Viewing replica logs for a model
Visualization in Cloudera AI quota management
Visualizations
Visualizing Experiment Results
Web session timeouts
What's New
What's New in ML Runtimes older releases
What's New in ML Runtimes version 2024.02.1
What's New in ML Runtimes version 2024.05.1
What's New in ML Runtimes version 2024.05.2
What's New in ML Runtimes version 2024.10.1
Workarounds for Cloudera Data Visualization with Hive and Impala
Workbench backup and restore prerequisites
Workbench editor file types
Worker network communication
Workers API
Workflows for active Models
Working with Data Discovery and Visualization
Yunikorn Gang scheduling

«

Filter topics

Securing Models

Cloudera AI Project Lifecycle
▶︎Managing Models
1. Models - Concepts and Terminology
▶︎Challenges with Machine Learning in production
▶︎Using Cloudera AI Registry
1. ▶︎Setting up Cloudera AI Registry
2. ▶︎Registering and deploying models with Cloudera AI Registry
▶︎Creating and deploying a Model
1. Hosting an LLM as a Cloudera AI Workbench model
2. Deploying the Cloudera AI Workbench 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
1. Access Keys for Models
2. ▶︎API Key for Models
Workflows for active Models
Technical metrics for Models
Debugging issues with Models
Deleting a Model
▶︎Example - Model training and deployment (Iris)
1. Training the Model
2. Deploying the Model

»

Securing Models

You can secure your Cloudera AI models using Access keys or API keys.

important

Cloudera on cloud allows customers to maintain full ownership and control of their data and workloads and is designed to operate in some of the most restricted on cloud environments. Since Cloudera on cloud runs in a customer’s cloud account, Security and Compliance is a shared responsibility between Cloudera and its on cloud customers. For more information, see Cloudera’s Shared Responsibility Model.

Related information

Cloudera’s Shared Responsibility Model

Access Keys for Models
Each model in Cloudera AI has a unique access key associated with it. This access key is a unique identifier for the model.
API Key for Models
You can prevent unauthorized access to your models by specifying an API key in the Authorization header of your model HTTP request. This topic covers how to create, test, and use an API key in Cloudera AI.

© 2019–2025 by Cloudera, Inc. All rights reserved.

We want your opinion

How can we improve this page?

What kind of feedback do you have?

Can we contact you for follow-up on this?

1.5
1.4
1. 1.4.1
2. 1.4.0