Custom quantization borders and missing value modes

    Contains

    Custom quantization borders and the method for processing missing values for the dataset numerical features.

    Format

    • Each line contains information regarding a single border and optionally the missing values mode settings for the corresponding feature.

    • Missing value modes are output for a feature if the following conditions are met at the same time:

      • The chosen missing value mode for the feature differs from Forbidden.
      • Missing values are present in the dataset.

      The global missing value mode is specified in the --nan-mode (nan_mode) training parameter and can be overridden in the Custom quantization borders and missing value modes input file.

    • Format of a single line:

      <zero-based feature ID><\t><border value><\t><missing value mode>
      

    Example

    0<\t>0.25
    0<\t>0.75
    2<\t>0.3<\t>Max
    2<\t>0.85<\t>Max