bob.machine.GMMMachine¶

class bob.machine.GMMMachine((object)self) → None¶

Bases: bob.machine._machine.MachineDoubleBase

This class implements a multivariate diagonal Gaussian distribution. See Section 2.3.9 of Bishop, “Pattern recognition and machine learning”, 2006

__init__( (object)self, (int)n_gaussians, (int)n_inputs) -> None

__init__( (object)self, (GMMMachine)other) -> None :: Creates a GMMMachine from another GMMMachine, using the copy constructor.

__init__( (object)self, (HDF5File)config) -> None

__init__((object)self) → None¶

__init__( (object)self, (int)n_gaussians, (int)n_inputs) -> None

__init__( (object)self, (GMMMachine)other) -> None :: Creates a GMMMachine from another GMMMachine, using the copy constructor.

__init__( (object)self, (HDF5File)config) -> None

Methods

`__init__`((object)self)	__init__( (object)self, (int)n_gaussians, (int)n_inputs) -> None
`acc_statistics`((GMMMachine)self, (object)x, …)	Accumulate the GMM statistics for this sample(s).
`acc_statistics_`((GMMMachine)self, (object)x, …)	Accumulate the GMM statistics for this sample(s).
`forward`((MachineDoubleBase)self, (object)input)	Executes the machine on the given 1D numpy array of float64, and returns the output.
`forward_`((MachineDoubleBase)self, (object)input)	Executes the machine on the given 1D numpy array of float64, and returns the output.
`is_similar_to`((GMMMachine)self, …)	Compares this GMMMachine with the ‘other’ one to be approximately the same.
`load`((GMMMachine)self, (HDF5File)config)	Load from a Configuration
`log_likelihood`((GMMMachine)self, (object)x, …)	Output the log likelihood of the sample, x, i.e.
`log_likelihood_`((GMMMachine)self, (object)x, …)	Output the log likelihood of the sample, x, i.e.
`resize`((GMMMachine)self, (int)n_gaussians, …)	Reset the input dimensionality, and the number of Gaussian components.
`save`((GMMMachine)self, (HDF5File)config)	Save to a Configuration
`set_variance_thresholds`((GMMMachine)self, …)	Set the variance flooring thresholds in each dimension to the same vector for all Gaussian components if the argument is a 1D numpy arrray, and equal for all Gaussian components and dimensions if the parameter is a scalar.
`update_gaussian`((GMMMachine)self, (int)i)	Get the specified Gaussian component.

Attributes

`dim_c`	The number of Gaussian components C
`dim_d`	The feature dimensionality D
`mean_supervector`	The mean supervector of the GMMMachine (concatenation of the mean vectors of each Gaussian of the GMMMachine
`means`	The means of the gaussians
`shape`	A tuple that represents the dimensionality of the GMMMachine `(n_gaussians, n_inputs)`.
`variance_supervector`	The variance supervector of the GMMMachine (concatenation of the variance vectors of each Gaussian of the GMMMachine
`variance_thresholds`	The variance flooring thresholds for each Gaussian in each dimension
`variances`	The (diagonal) variances of the Gaussians
`weights`	The weights (also known as “mixing coefficients”)

__call__((MachineDoubleBase)self, (object)input) → float :¶: Executes the machine on the given 1D numpy array of float64, and returns the output. NO CHECK is performed.

acc_statistics((GMMMachine)self, (object)x, (GMMStats)stats) → None :¶: Accumulate the GMM statistics for this sample(s). Inputs are checked.

acc_statistics_((GMMMachine)self, (object)x, (GMMStats)stats) → None :¶: Accumulate the GMM statistics for this sample(s). Inputs are NOT checked.

forward((MachineDoubleBase)self, (object)input) → float :¶: Executes the machine on the given 1D numpy array of float64, and returns the output.

forward_((MachineDoubleBase)self, (object)input) → float :¶: Executes the machine on the given 1D numpy array of float64, and returns the output. NO CHECK is performed.

is_similar_to((GMMMachine)self, (GMMMachine)other[, (float)r_epsilon=1e-05[, (float)a_epsilon=1e-08]]) → bool :¶: Compares this GMMMachine with the ‘other’ one to be approximately the same.

load((GMMMachine)self, (HDF5File)config) → None :¶: Load from a Configuration

log_likelihood((GMMMachine)self, (object)x, (object)log_weighted_gaussian_likelihoods) → float :¶

Output the log likelihood of the sample, x, i.e. log(p(x|bob::machine::GMMMachine)). Inputs are checked.

log_likelihood( (GMMMachine)self, (object)x) -> float :: Output the log likelihood of the sample, x, i.e. log(p(x|GMM)). Inputs are checked.

log_likelihood_((GMMMachine)self, (object)x, (object)log_weighted_gaussian_likelihoods) → float :¶

Output the log likelihood of the sample, x, i.e. log(p(x|bob::machine::GMMMachine)). Inputs are NOT checked.

log_likelihood_( (GMMMachine)self, (object)x) -> float :: Output the log likelihood of the sample, x, i.e. log(p(x|GMM)). Inputs are checked.

mean_supervector¶: The mean supervector of the GMMMachine (concatenation of the mean vectors of each Gaussian of the GMMMachine

resize((GMMMachine)self, (int)n_gaussians, (int)n_inputs) → None :¶: Reset the input dimensionality, and the number of Gaussian components. Initialises the weights to uniform distribution.

save((GMMMachine)self, (HDF5File)config) → None :¶: Save to a Configuration

set_variance_thresholds((GMMMachine)self, (object)variance_threshold) → None :¶: Set the variance flooring thresholds in each dimension to the same vector for all Gaussian components if the argument is a 1D numpy arrray, and equal for all Gaussian components and dimensions if the parameter is a scalar.

shape¶: A tuple that represents the dimensionality of the GMMMachine (n_gaussians, n_inputs).

update_gaussian((GMMMachine)self, (int)i) → Gaussian :¶: Get the specified Gaussian component. An exception is thrown if i is out of range.

variance_supervector¶: The variance supervector of the GMMMachine (concatenation of the variance vectors of each Gaussian of the GMMMachine

variance_thresholds¶: The variance flooring thresholds for each Gaussian in each dimension