Before You Begin

• Sampling in Driverless AI
  • Data Sampling
  • Imbalanced Model Sampling Methods
• Missing and Unseen Levels Handling
  • How Does the Algorithm Handle Missing Values During Training?
  • How Does the Algorithm Handle Missing Values During Scoring (Production)?
  • What Happens When You Try to Predict on a Categorical Level Not Seen During Training?
  • What Happens if the Response Has Missing Values?
• Imputation in Driverless AI
  • Enabling Imputation
  • Running an Experiment with Imputation
• Driverless AI Transformations
  • Available Transformers
  • Transformed Feature Naming Convention
  • Example Transformations
• Internal Validation Technique
• Ensemble Learning in Driverless AI
  • Ensemble Method
  • Ensemble Levels
• Monotonicity Constraints
• Variable importance in Driverless AI
  • Global Feature Importance
  • Local Feature Importance
• Wide Datasets in Driverless AI
• GPUs in Driverless AI
• Experiment Queuing In Driverless AI
• Time Series Best Practices
  • Preparing Your Data
  • Experiment Setup
  • Interpreting Models with MLI
  • Scoring
  • Other Approaches
• Tips ‘n Tricks
  • Pipeline Tips
  • Time Series Tips
  • Scorer Tips
  • Knob Settings Tips
  • Tips for Running an Experiment
  • Expert Settings Tips
  • Checkpointing Tips
  • Text Data Tips
• Simple Configurations
  • Get a quick Final Model: no Genetic Algorithm no Ensembling
  • Use Original Features With Genetic Algorithm
  • Build models with your choice of algorithm and parameters
  • Disable leakage checks, shift detection, mojo creation and ensembling