sksurv.kernels.ClinicalKernelTransform

class sksurv.kernels.ClinicalKernelTransform(fit_once=False, _numeric_ranges=None, _numeric_columns=None, _nominal_columns=None)[source]

Transform data using a clinical Kernel

The clinical kernel distinguishes between continuous ordinal,and nominal variables.

See [1] for further description.

Parameters:fit_once (bool, optional) – If set to True, fit() does only transform the training data, but not update its internal state. You should call prepare() once before calling transform(). If set to False, it behaves like a regular estimator, i.e., you need to call fit() before transform().

References

[1]Daemen, A., De Moor, B., “Development of a kernel function for clinical data”. Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 5913-7, 2009
__init__(fit_once=False, _numeric_ranges=None, _numeric_columns=None, _nominal_columns=None)[source]

Initialize self. See help(type(self)) for accurate signature.

Methods

__init__([fit_once, _numeric_ranges, …]) Initialize self.
fit(X[, y]) Determine transformation parameters from data in X.
pairwise_kernel(X, Y) Function to use with sklearn.metrics.pairwise.pairwise_kernels()
prepare(X) Determine transformation parameters from data in X.
transform(Y) Compute all pairwise distances between self.X_fit_ and Y.
fit(X, y=None, **kwargs)[source]

Determine transformation parameters from data in X.

Subsequent calls to transform(Y) compute the pairwise distance to X. Parameters of the clinical kernel are only updated if fit_once is False, otherwise you have to explicitly call prepare() once.

Parameters:
  • X (pandas.DataFrame, shape = (n_samples, n_features)) – Data to estimate parameters from.
  • y (None) – Argument is ignored (included for compatibility reasons).
  • kwargs (dict) – Argument is ignored (included for compatibility reasons).
Returns:

self – Returns the instance itself.

Return type:

object

pairwise_kernel(X, Y)[source]

Function to use with sklearn.metrics.pairwise.pairwise_kernels()

Parameters:
  • X (array, shape = (n_features,)) –
  • Y (array, shape = (n_features,)) –
Returns:

similarity – Similarities are normalized to be within [0, 1]

Return type:

float

prepare(X)[source]

Determine transformation parameters from data in X.

Use if fit_once is True, in which case fit() does not set the parameters of the clinical kernel.

Parameters:X (pandas.DataFrame, shape = (n_samples, n_features)) – Data to estimate parameters from.
transform(Y)[source]

Compute all pairwise distances between self.X_fit_ and Y.

Parameters:y (array-like, shape = (n_samples_y, n_features)) –
Returns:kernel – Kernel matrix. Values are normalized to lie within [0, 1].
Return type:ndarray, shape = (n_samples_y, n_samples_X_fit_)