Sklearn Standard Scaler icon

Sklearn Standard Scaler

Standardize features using scikit-learn StandardScaler

Actions3

Transform

AI-generated

Summary

Scales specified feature columns using a provided fitted StandardScaler and adds scaled columns with a custom prefix.

Inputs

  • scalerData (required) — JSON string containing the fitted scaler (mean, scale, with_mean, with_std) from a prior Fit operation.
  • transformFeatureColumns (required) — Comma-separated list of column names to scale; must match those used during fitting.
  • outputPrefix — Prefix added to the names of the scaled columns. Defaults to 'scaled_'.
  • pythonPath — Path to the Python executable that has scikit-learn installed. Defaults to 'python3'.

Output shape

Each input item is returned unchanged plus new columns named for each specified feature, containing the standardized (zero‑mean, unit‑variance) values.

If the scaler data lacks mean or scale (e.g., with_mean or with_std is false), the corresponding dimension is left unchanged (treated as offset 0 or factor 1).

Examples

Example 1: Transform numeric columns 'age' and 'salary' using a previously fitted scaler.

Set scalerData to the JSON from the Fit operation, transformFeatureColumns to 'age,salary', outputPrefix to 'scaled_' (default).

Discussion