yobx.sklearn.decomposition.gaussian_random_projection#

yobx.sklearn.decomposition.gaussian_random_projection.sklearn_gaussian_random_projection(g: GraphBuilderExtendedProtocol, sts: Dict, outputs: List[str], estimator: GaussianRandomProjection, X: str, name: str = 'gaussian_random_projection') → str[source]#

Converts a sklearn.random_projection.GaussianRandomProjection into ONNX.

The projection simply multiplies the input by the transposed random projection matrix stored in components_:

X  ──MatMul(components_.T)──►  output
Parameters:

  • g – the graph builder to add nodes to

  • sts – shapes defined by scikit-learn

  • outputs – desired output names (projected inputs)

  • estimator – a fitted GaussianRandomProjection

  • X – input tensor name

  • name – prefix name for the added nodes

Returns:

output tensor name