gw.SVD.SVDBasis

gw.SVD.SVDBasis(file_name=None, dictionary=None)

Attributes

Name	Description
compress	Convert from data (e.g., frequency series) to compressed representation in
compute_test_mismatches	Test SVD basis by computing mismatches of compressed / decompressed data
decompress	Convert from basis coefficients back to raw data representation.
from_dictionary	Load the SVD basis from a dictionary.
from_file	Load the SVD basis from a HDF5 file.
generate_basis	Generate the SVD basis from training data and store it.
print_validation_summary	Print a summary of the validation mismatches.

gw.SVD.SVDBasis.compress(data)

Convert from data (e.g., frequency series) to compressed representation in terms of basis coefficients.

Name	Type	Description	Default
data	np.ndarray		required

Name	Type	Description
	array of basis coefficients

gw.SVD.SVDBasis.compute_test_mismatches(
    data,
    parameters=None,
    increment=50,
    verbose=False,
)

Test SVD basis by computing mismatches of compressed / decompressed data against original data. Results are saved as a DataFrame.

Name	Type	Description	Default
data	np.ndarray	Array of data sets to validate against.	required
parameters	pd.DataFrame	Optional labels for the data sets. This is useful for checking performance on particular regions of the parameter space.	`None`
increment	int	Specifies SVD truncations for computing mismatches. E.g., increment = 50 means that the SVD will be truncated at size [50, 100, 150, …, len(data)].	`50`
verbose	bool	Whether to print summary statistics.	`False`

gw.SVD.SVDBasis.decompress(coefficients)

Convert from basis coefficients back to raw data representation.

Name	Type	Description	Default
coefficients	np.ndarray	Array of basis coefficients	required

Name	Type	Description
	array of decompressed data

gw.SVD.SVDBasis.from_dictionary(dictionary)

Load the SVD basis from a dictionary.

Name	Type	Description	Default
dictionary	dict	The dictionary should contain at least a ‘V’ key, and optionally an ‘s’ key.	required

gw.SVD.SVDBasis.from_file(filename)

Load the SVD basis from a HDF5 file.

Name	Type	Description	Default
filename	str		required

gw.SVD.SVDBasis.generate_basis(training_data, n, method='scipy')

Generate the SVD basis from training data and store it.

The SVD decomposition takes

training_data = U @ diag(s) @ Vh

where U and Vh are unitary.

Name	Type	Description	Default
training_data	np.ndarray	Array of waveform data on the physical domain	required
n	int	Number of basis elements to keep. n=0 keeps all basis elements.	required
method	str	Select SVD method, ‘random’ or ‘scipy’	`'scipy'`

gw.SVD.SVDBasis.print_validation_summary()

Print a summary of the validation mismatches.