Bring Your Own Recipes

Data scientists can extend the Driverless AI platform by uploading their own models, transformers and scorers as a custom recipe. Bring-Your-Own recipes or use the examples built in the open and curated by the data science community. Driverless AI treats recipes as first-class citizens in the automatic machine learning workflow.

Machine Learning Interpretability

H2O Driverless AI provides robust interpretability of machine learning models to explain modeling results. In the MLI view H2O Driverless AI employs a host of different techniques and methodologies for interpreting and explaining the results of its models, four charts are generated automatically including: K-LIME, Shapley, Variable Importance, Decision Tree, Partial Dependence and more.

Natural Language Processing

Text data can contain critical information to inform better predictions. Driverless AI automatically converts text strings into features using powerful techniques like TFIDF, CNN, and GRU. With TensorFlow, Driverless AI can also process larger text blocks and build models using all available data and to solve business problems like sentiment analysis, document classification and content tagging.

Flexibility of Data Ingestion and Compute Technologies

H2O Driverless AI can ingest data from a variety of data sets including Hadoop HDFS, Amazon S3, and more. H2O Driverless AI can also be deployed everywhere including all clouds (Microsoft Azure, AWS, Google Cloud) and on-premises on any systems.

H2O Driverless AI is optimized to work with the with the latest Nvidia GPUs, IBM Power 9 and Intel x86 CPUs and to take advantage of GPU acceleration to achieve up to 30X speedups for automatic machine learning. Driverless AI includes support for GPU accelerated algorithms like XGBoost, TensorFlow, LightGBM GLM, and more.