Sklearn MinMax Scaler icon

Sklearn MinMax Scaler

Scale features to a given range using scikit-learn MinMaxScaler

Actions4

Inverse Transform

AI-generated

Summary

Undo a MinMax scaling applied to selected feature columns by using a previously fitted scaler.

Inputs

  • scalerData (required) — JSON string containing the fitted scaler (from fit operation)
  • transformFeatureColumns (required) — Comma-separated list of feature column names to inverse-transform
  • outputPrefix — Prefix to add to scaled column names
  • pythonPath — Path to Python executable with scikit-learn installed

Output shape

a list of items

Each returned item retains the original fields and adds new columns prefixed by outputPrefix (default "scaled_") with the inverse‑transformed values; failure may produce an error object if continueOnFail is disabled.

Examples

Example 1: Undo scaling on columns a and b

operation: Inverse Transform; scalerData: '{"data_min":[0,0],"data_max":[1,1],"feature_range":[0,1],"feature_columns":["a","b"]}' ; transformFeatureColumns: "a,b" ; outputPrefix: "scaled_"

Example 2: Inverse transform with custom prefix

operation: Inverse Transform; scalerData: '{"data_min":[0,0],"data_max":[1,1],"feature_range":[0,1],"feature_columns":["x","y"]}' ; transformFeatureColumns: "x,y" ; outputPrefix: "inv_"

Example 3: Inverse transform using a specific Python interpreter

operation: Inverse Transform; scalerData: '{"data_min":[0,0],"data_max":[1,1],"feature_range":[0,1],"feature_columns":["c","d"]}' ; transformFeatureColumns: "c,d" ; outputPrefix: "scaled_" ; pythonPath: "/usr/bin/python3.9"

Discussion