hummingbird.ml.operator_converters.sklearn.decomposition

Converters for scikit-learn matrix and cross decomposition transformers: PCA, KernelPCA, TruncatedSVD, FastICA.

hummingbird.ml.operator_converters.sklearn.decomposition.convert_sklearn_fast_ica(operator, device, extra_config)

Converter for sklearn.decomposition.FastICA

Args:

operator: An operator wrapping a sklearn.decomposition.FastICA transformer device: String defining the type of device the converted operator should be run on extra_config: Extra configuration used to select the best conversion strategy

Returns:

A PyTorch model

hummingbird.ml.operator_converters.sklearn.decomposition.convert_sklearn_kernel_pca(operator, device, extra_config)

Converter for sklearn.decomposition.KernelPCA

Args:

operator: An operator wrapping a sklearn.decomposition.KernelPCA transformer device: String defining the type of device the converted operator should be run on extra_config: Extra configuration used to select the best conversion strategy

Returns:

A PyTorch model

hummingbird.ml.operator_converters.sklearn.decomposition.convert_sklearn_pca(operator, device, extra_config)

Converter for sklearn.decomposition.PCA

Args:

operator: An operator wrapping a sklearn.decomposition.PCA transformer device: String defining the type of device the converted operator should be run on extra_config: Extra configuration used to select the best conversion strategy

Returns:

A PyTorch model

hummingbird.ml.operator_converters.sklearn.decomposition.convert_sklearn_pls_regression(operator, device, extra_config)

Converter for sklearn.cross_decomposition.PLSRegression

Args:

operator: An operator wrapping a sklearn.cross_decomposition.PLSRegression transformer device: String defining the type of device the converted operator should be run on extra_config: Extra configuration used to select the best conversion strategy

Returns:

A PyTorch model

hummingbird.ml.operator_converters.sklearn.decomposition.convert_sklearn_truncated_svd(operator, device, extra_config)

Converter for sklearn.decomposition.TruncatedSVD

Args:

operator: An operator wrapping a sklearn.decomposition.TruncatedSVD transformer device: String defining the type of device the converted operator should be run on extra_config: Extra configuration used to select the best conversion strategy

Returns:

A PyTorch model