Package gmisclib :: Module multivariance_q :: Class diag_quadratic_with_numbers

Class diag_quadratic_with_numbers

Instance Methods

__init__(self, mu, invsigma, details, bias=0.0, offset=None)
Bias is an overall shift of the log probability up or down.

source code

__str__(self)

source code

__repr__(self)

source code

source code

logp(self, datum)

source code

ndim(self) (Inherited from gmisclib.multivariance_classes.model_with_numbers)

source code

new(self, mu, invsigma) (Inherited from gmisclib.multivariance_classes.model_with_numbers)

source code

offset(self) (Inherited from gmisclib.multivariance_classes.model_with_numbers)

source code

pack(self)
Returns a vector of parameters. (Inherited from gmisclib.multivariance_classes.model_with_numbers)

source code

start(self, dataset) (Inherited from gmisclib.multivariance_classes.model_with_numbers)

source code

unpack(self, prms) (Inherited from gmisclib.multivariance_classes.model_with_numbers)

source code

Class Variables
	__doc__ = `"""Virtual base class for adding in the functi...` (Inherited from gmisclib.multivariance_classes.model_with_numbers)

Method Details

Bias is an overall shift of the log probability up or down. In a classifier, it is used to bias things toward one class or another.

Overrides: multivariance_classes.model_with_numbers.__init__: (inherited documentation)