What Are AI Recipes?

  • Recipes are customizations and extensions to the Driverless AI platform.
  • Data scientists can bring their own recipes or leverage the ones built by H2O.ai’s data science community.
  • Recipes can be any one or combination of the following:
    • Custom transformers
    • Custom machine learning models
    • Custom scorers (classification or regression)
    • Custom datasets

Driverless AI + Your Recipes = A Truly Extensible AI Platform

  • Flexibility, extensibility and customizations built into the Driverless AI platform
  • New recipes built by the data science community, curated by Kaggle Grand Masters @ H2o.ai
  • Data scientists can focus on domain-specific functions to build customizations.
  • 1-click upload of your recipes – algorithms, scorers and transformations
  • Driverless AI treats custom recipes as first-class citizens in the automatic machine learning workflow.
  • Every business can have a recipe cookbook for collaborative data science with their organization

How Driverless AI Recipes Work

Examples of Recipes

  • Models
  • Scorers
  • Transformers
  • Datasets

CatBoost, H2O-3 Models, k-Nearest Neighbor, Linear Support Vector Machine (SVM), Light Gradient Boosting Machine (GBM), XGBoost, and Historic Mean.

False Discovery Rate, Hamming Loss, Hyperbolic Cosine Loss, Mean Absolute Scaled Error for time-series regression, Pearson Correlation Coefficient for regression, and Huber Loss for Regression or Classification.

Sentiment extraction from text, Exponentiated difference of two numbers, Anomaly score based on deep learning autoencoder, Historical volatility calculator and Text similarity based on FuzzyWuzzy.

Alert if a date falls on one of Singapore Public Holidays, Ramadan days in Saudi Arabia, or any of the Germany Bundesland holidays.

Driverless AI Screen
For additional examples of Recipes, please visit: https://github.com/h2oai/driverlessai-recipes/