• Documentation
    • Search documentation
    karat

    +

    K

    API Reference ↗
    Data connectivity & integrationPipeline Builder TransformsDimensionality reduction

    Dimensionality reduction

    Supported in: Batch

    Reduces dimensionality of an n-dimensional array to a k-dimensional array by applying principal component analysis.

    Transform categories: Other

    Declared arguments

    • Column - Column with arrays on which dimensionality reduction will be applied.
      Column<Embedded vector | Embedded vector>
    • Dataset - Dataset to apply dimensionality reduction to.
      Table
    • optional Output dimensions - Dimension of the output array after reduction is applied.
      Literal<Integer>

    Examples

    Example 1: Base case

    Argument values:

    • Column: array
    • Dataset: ri.foundry.main.dataset.a
    • Output dimensions: 2

    Input:

    array
    [ 1.0, 2.0, 3.0 ]
    [ 4.0, 5.0, 6.0 ]
    [ 7.0, 8.0, 9.0 ]
    [ 10.0, 11.0, 12.0 ]

    Output:

    arrayarray_pca
    [ 1.0, 2.0, 3.0 ][ -3.46, 1.36 ]
    [ 4.0, 5.0, 6.0 ][ -8.66, 1.36 ]
    [ 7.0, 8.0, 9.0 ][ -13.85, 1.36 ]
    [ 10.0, 11.0, 12.0 ][ -19.05, 1.36 ]

    Example 2: Base case

    Argument values:

    • Column: array
    • Dataset: ri.foundry.main.dataset.a
    • Output dimensions: 3

    Input:

    array
    [ 1.0, 2.0, 3.0, 4.0, 5.0 ]
    [ 6.0, 7.0, 8.0, 9.0, 10.0 ]
    [ 11.0, 12.0, 13.0, 14.0, 15.0 ]
    [ 16.0, 17.0, 18.0, 19.0, 20.0 ]

    Output:

    arrayarray_pca
    [ 1.0, 2.0, 3.0, 4.0, 5.0 ][ -6.71, -2.24, -1.73 ]
    [ 6.0, 7.0, 8.0, 9.0, 10.0 ][ -17.89, -2.24, -1.73 ]
    [ 11.0, 12.0, 13.0, 14.0, 15.0 ][ -29.07, -2.24, -1.73 ]
    [ 16.0, 17.0, 18.0, 19.0, 20.0 ][ -40.25, -2.24, -1.73 ]

    Example 3: Null case

    Argument values:

    • Column: array
    • Dataset: ri.foundry.main.dataset.a
    • Output dimensions: 2

    Input:

    array
    [ 1.0, null ]
    [ null, null ]
    [ null, 1.0 ]

    Output:

    arrayarray_pca
    [ 1.0, null ][ -0.7, 0.7 ]
    [ null, null ][ 0.0, 0.0 ]
    [ null, 1.0 ][ 0.7, 0.7 ]

    Example 4: Null case

    Argument values:

    • Column: array
    • Dataset: ri.foundry.main.dataset.a
    • Output dimensions: 2

    Input:

    array
    [ 1.0, null ]
    null
    [ null, 1.0 ]

    Output:

    arrayarray_pca
    [ 1.0, null ][ -0.7, 0.7 ]
    [ null, 1.0 ][ 0.7, 0.7 ]