H2O Driverless AI is an artificial intelligence (AI) platform for automatic machine learning. Driverless AI automates some of the most difficult data science and machine learning workflows such as feature engineering, model validation, model tuning, model selection, and model deployment. It aims to achieve highest predictive accuracy, comparable to expert data scientists, but in much shorter time thanks to end-to-end automation. Driverless AI also offers automatic visualizations and machine learning interpretability (MLI). Especially in regulated industries, model transparency and explanation are just as important as predictive performance. Modeling pipelines (feature engineering and models) are exported (in full fidelity, without approximations) both as Python modules and as Java standalone scoring artifacts.
Driverless AI runs on commodity hardware. It was also specifically designed to take advantage of graphical processing units (GPUs), including multi-GPU workstations and servers such as IBM’s Power9-GPU AC922 server and the NVIDIA DGX-1 for order-of-magnitude faster training.
This document describes how to install and use Driverless AI. For more information about Driverless AI, see https://www.h2o.ai/products/h2o-driverless-ai/.
For a third-party review, see https://www.infoworld.com/article/3236048/machine-learning/review-h2oai-automates-machine-learning.html.
If you have questions about using Driverless AI, post them on Stack Overflow using the driverless-ai tag at http://stackoverflow.com/questions/tagged/driverless-ai. You can also post questions on the H2O.ai Community Slack workspace in the #driverlessai channel. If you have not signed up for the H2O.ai Community Slack workspace, you can do so here: https://www.h2o.ai/community/.
- Why Driverless AI?
- Key Features
- Flexibility of Data and Deployment
- NVIDIA GPU Acceleration
- Automatic Data Visualization (Autovis)
- Automatic Feature Engineering
- Automatic Model Documentation
- Time Series Forecasting
- NLP with TensorFlow
- Automatic Scoring Pipelines
- Machine Learning Interpretability (MLI)
- Automatic Reason Codes
- Bring Your Own Recipe (BYOR) Support
- Supported Algorithms
- Driverless AI Workflow
- Supported Environments
- Before You Begin Installing or Upgrading
- Supported Browsers
sudoor Not to
- Note about nvidia-docker 1.0
- Deprecation of
nvidia-container-runtime-hookRequirement for PowerPC Users
- Note About CUDA Versions
- Note About Authentication
- Note About Shared File Systems
- Note About the Master Database File
- Backup Strategy
- Upgrade Strategy
- Sizing Requirements
- Installing and Upgrading Driverless AI
- Using the config.toml File
- Environment Variables and Configuration Options
- Enabling Data Connectors
- Configuring Authentication
- Enabling Notifications
- Export Artifacts
- Changing the Language in the UI
- Launching Driverless AI
- The Datasets Page
- Diagnosing a Model
- Project Workspace
- Scoring Pipelines Overview
- Visualizing the Scoring Pipeline
- Which Pipeline Should I Use?
- Driverless AI Standalone Python Scoring Pipeline
- Driverless AI MLI Standalone Python Scoring Package
- MOJO Scoring Pipelines
- What’s Happening in Driverless AI?
- Data Sampling
- Driverless AI Transformations
- Internal Validation Technique
- Missing and Unseen Levels Handling
- Imputation in Driverless AI
- Time Series in Driverless AI
- NLP in Driverless AI
- The Python Client
- The R Client
- Tips ‘n Tricks