Singular Value Decomposition

This Component uses Singular Value Decomposition to transform given numeric columns and appends the components as new columns to the input table. Additionally, the Component provides a table containing the explained variance ratio and the fitted model.

The Component uses the Python Extension to perform the SVD with the Python Class “Dimensionality reduction using truncated SVD (aka LSA)” in the sci-kit learn library (https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.TruncatedSVD.html).

Input ports

1. Type: Table
   Input table
   The table that contains numeric columns to be decomposed.

Output ports

1. Type: Table
   Result table
   The resulting table containing the original columns and the single components.
2. Type: Table
   Explained Variance Ratio
   Table that contains the explained variance ratio per component.
3. Type: Python
   Pickled Object
   The pickled object that can be used to transform other tables.