Sklearn Truncated SVD icon

Sklearn Truncated SVD

Dimensionality reduction using truncated SVD (LSA) for sparse data

Actions2

Transform

AI-generated

Summary

Transform incoming data using a pre‑fitted Truncated SVD model.

Inputs

  • svdData (required) — JSON string containing the fitted SVD model
  • transformFeatureColumns (required) — Comma‑separated list of feature column names to transform
  • pythonPath — Path to the Python executable that has scikit‑learn installed (default: python3)

Output shape

a list of items extending each input record with SVD component columns and metadata fields

Adds fields svd_component_1, svd_component_2, … and copies metadata such as explained_variance_ratio and total_variance_explained; may return an error envelope on failure unless continueOnFail is enabled.

Examples

Example 1: Transform a dataset using a previously fitted SVD model stored in a JSON string

Provide svdData with the fitted model JSON, list the feature columns in transformFeatureColumns, optionally set pythonPath if using a custom environment.

Example 2: Add SVD components to each row of a CSV‑imported table

Set svdData to the saved model JSON, list the column names in transformFeatureColumns, keep default pythonPath.

Discussion