Binner

Groups numeric columns into discrete intervals, known as bins. This node supports both automatic and manual binning methods, including equal-width, equal-frequency, custom cutoffs, and quantile-based binning. Each bin is assigned a label based on its position, boundary values, or midpoint. The resulting binned values are output as string-type columns, either replacing the original data or appended alongside it.

Options

Columns to bin
Select the numeric columns to apply binning to. Only the selected columns will be transformed.
Binning type
Select the method used to define bin intervals.
  • Equal width: Creates bins of equal size across the value range. Requires a specified number of bins.
  • Equal frequency: Creates bins with approximately the same number of values. Requires a specified number of bins.
  • Custom cutoffs: Manually define bin edges using a list of cutoff values. At least two cutoffs are required.
  • Custom quantiles: Define bin edges based on quantile values. At least two quantiles are required. Quantiles are converted to bin edges using the R-7 algorithm, see WP:Quantile for more details. Note that when setting upper or lower bounds quantiles are calculated only on values within those bounds.
Number of bins
Specifies the number of bins to create. Used with Equal width and Equal frequency binning.
Custom cutoffs
Specifies the exact value at which the bin boundary should be placed.
  • Cutoff: Specifies the exact value at which the bin boundary should be placed.
  • Exact match: Defines how to assign values that match a cutoff exactly.
    • To lower bin: Assign to the bin below the cutoff.
    • To upper bin: Assign to the bin above the cutoff.
Custom quantiles
The bin boundaries may be defined manually by entering the quantiles. The behaviour when a value is exactly equal to a quantile can be configured separately for each quantile.
  • Quantile: Specifies a quantile value (between 0 and 1) used as a bin boundary.
  • Exact match: Defines how to assign values that match a cutoff exactly.
    • To lower bin: Assign to the bin below the cutoff.
    • To upper bin: Assign to the bin above the cutoff.
Enforce integer cutoffs
If enabled, bin boundaries will be rounded to the nearest integer. Not available when using custom cutoffs.
Fix lower bound
If enabled, values below the specified lower bound will be assigned to a dedicated outlier bin. The bin will have the name specified by the 'Lower outlier value' setting.
Lower bound
Sets the minimum value for binning. Values below this will be assigned to the lower outlier bin.
Fix upper bound
If enabled, values above the specified upper bound will be assigned to a dedicated outlier bin. The bin will have the name specified by the 'Upper outlier value' setting.
Upper bound
Sets the maximum value for binning. Values above this will be assigned to the upper outlier bin.
Bin names/values
Selects how bins are labeled in the output. Use a 'Number Format Manager' node to customize number formatting.
  • Numbered (e.g., Bin 1, Bin 2): Bins are labeled by index.
  • Borders (e.g., [0.0, 1.0)): Bins are labeled using interval borders.
  • Midpoints: Bins are labeled using the midpoint of each interval.
Prefix
Text to prepend to bin labels when using numbered naming (e.g., Bin 1, Bin 2).
Lower outlier value
Label assigned to values below the fixed lower bound.
Upper outlier value
Label assigned to values above the fixed upper bound.
Output columns
Whether to replace the original columns with the binned columns, or to append the binned columns with new names created by appending the specified suffix to the original column names.
  • Replace: Replaces the original column with the binned result.
  • Append with suffix: Adds a new column with the binned values.
Suffix
Text appended to column names when new columns are created using the Append option.

Input Ports

Icon
Data to be categorized

Output Ports

Icon
Data with bins defined
Icon
The PMML Model fragment containing information how to bin

Popular Predecessors

  • No recommendations found

Popular Successors

  • No recommendations found

Views

This node has no views

Workflows

  • No workflows found

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.