grakel.HadamardCode

class grakel.HadamardCode(n_jobs=None, verbose=False, normalize=False, n_iter=5, base_graph_kernel=None)

The simple Hadamard code kernel, as proposed in [KI16].

Parameters:

base_graph_kernelgrakel.kernels.Kernel or tuple, default=None

If tuple it must consist of a valid kernel object and a dictionary of parameters. General parameters concerning normalization, concurrency, .. will be ignored, and the ones of given on __init__ will be passed in case it is needed. Default base_graph_kernel is VertexHistogram.

rhoint, condition_of_appearance: hc_type==”shortened”, default=-1

The size of each single bit arrays. If -1 is chosen r is calculated as the biggest possible that satisfies an equal division.

Lint, condition_of_appearance: hc_type==”shortened”, default=4

The number of bytes to store the bitarray of each label.

n_iterint, default=5

The number of iterations.

Attributes:

base_graph_kernel_function

A void function that initializes a base kernel object.

Methods

diagonal()	Calculate the kernel matrix diagonal for fitted data.
fit(X[, y])	Fit a dataset, for a transformer.
fit_transform(X[, y])	Fit and transform, on the same dataset.
get_metadata_routing()	Get metadata routing of this object.
get_params([deep])	Get parameters for this estimator.
initialize()	Initialize all transformer arguments, needing initialization.
pairwise_operation(x, y)	Calculate a pairwise kernel between two elements.
parse_input(X)	Parse input and create features, while initializing and/or calculating sub-kernels.
set_output(*[, transform])	Set output container.
set_params(**params)	Call the parent method.
transform(X)	Calculate the kernel matrix, between given and fitted dataset.

Initialise a hadamard_code kernel.

Attributes:

X

Methods

diagonal()	Calculate the kernel matrix diagonal for fitted data.
fit(X[, y])	Fit a dataset, for a transformer.
fit_transform(X[, y])	Fit and transform, on the same dataset.
get_metadata_routing()	Get metadata routing of this object.
get_params([deep])	Get parameters for this estimator.
initialize()	Initialize all transformer arguments, needing initialization.
pairwise_operation(x, y)	Calculate a pairwise kernel between two elements.
parse_input(X)	Parse input and create features, while initializing and/or calculating sub-kernels.
set_output(*[, transform])	Set output container.
set_params(**params)	Call the parent method.
transform(X)	Calculate the kernel matrix, between given and fitted dataset.

__init__(n_jobs=None, verbose=False, normalize=False, n_iter=5, base_graph_kernel=None)

Initialise a hadamard_code kernel.

Bibliography

[KI16]

Tetsuya Kataoka and Akihiro Inokuchi. Hadamard code graph kernels for classifying graphs. In Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods, 24–32. 2016.