bob.trainer.PLDATrainer¶

class bob.trainer.PLDATrainer((object)arg1) → None¶

Bases: bob.trainer._trainer.EMTrainerPLDA

A trainer for Probabilistic Linear Discriminant Analysis (PLDA). The train() method will learn the mu, F, G and Sigma of the model, whereas the enrol() method, will store model information about the enrolment samples for a specific class.

References: 1. ‘A Scalable Formulation of Probabilistic Linear Discriminant Analysis: Applied to Face Recognition’, Laurent El Shafey, Chris McCool, Roy Wallace, Sebastien Marcel, TPAMI‘2013 2. ‘Probabilistic Linear Discriminant Analysis for Inference About Identity’, Prince and Elder, ICCV‘2007. 3. ‘Probabilistic Models for Inference about Identity’, Li, Fu, Mohammed, Elder and Prince, TPAMI‘2012.

__init__( (object)self [, (int)max_iterations=100 [, (bool)use_sum_second_order=True]]) -> None :: Initializes a new PLDATrainer.
__init__( (object)self, (PLDATrainer)trainer) -> None :: Copy constructs a PLDATrainer

__init__((object)arg1) → None¶

__init__( (object)self [, (int)max_iterations=100 [, (bool)use_sum_second_order=True]]) -> None :: Initializes a new PLDATrainer.
__init__( (object)self, (PLDATrainer)trainer) -> None :: Copy constructs a PLDATrainer

Methods

`__init__`((object)arg1)	__init__( (object)self [, (int)max_iterations=100 [, (bool)use_sum_second_order=True]]) -> None :
`e_step`((EMTrainerPLDA)self, …)	Updates the hidden variable distribution (or the sufficient statistics) given the Machine parameters.
`enrol`((PLDATrainer)self, …)	Enrol a class-specific model (PLDAMachine) given a set of enrolment samples.
`finalize`((EMTrainerPLDA)self, …)	This method is called at the end of the EM algorithm
`initialize`((EMTrainerPLDA)self, …)	This method is called before the EM algorithm
`is_similar_to`((PLDATrainer)self, …)	Compares this PLDATrainer with the ‘other’ one to be approximately the same.
`m_step`((EMTrainerPLDA)self, …)	Updates the Machine parameters given the hidden variable distribution (or the sufficient statistics)
`train`((EMTrainerPLDA)self, …)	Trains a PLDABase using data (mu, F, G and sigma are learnt).

Attributes

`BETWEEN_SCATTER`
`CONSTANT`
`RANDOM_F`
`RANDOM_G`
`RANDOM_SIGMA`
`VARIANCE_DATA`
`VARIANCE_G`
`WITHIN_SCATTER`
`init_f_method`	The method used for the initialization of F.
`init_f_ratio`	The ratio used for the initialization of F.
`init_g_method`	The method used for the initialization of G.
`init_g_ratio`	The ratio used for the initialization of G.
`init_sigma_method`	The method used for the initialization of sigma.
`init_sigma_ratio`	The ratio used for the initialization of sigma.
`max_iterations`	Max iterations
`rng`	The Mersenne Twister mt19937 random generator used for the initialization of subspaces/arrays before the EM loop.
`use_sum_second_order`	Tells whether the second order statistics are stored during the training procedure, or only their sum.
`z_first_order`
`z_second_order`
`z_second_order_sum`

BETWEEN_SCATTER = bob.trainer._trainer.init_f_method.BETWEEN_SCATTER¶

CONSTANT = bob.trainer._trainer.init_sigma_method.CONSTANT¶

RANDOM_F = bob.trainer._trainer.init_f_method.RANDOM_F¶

RANDOM_G = bob.trainer._trainer.init_g_method.RANDOM_G¶

RANDOM_SIGMA = bob.trainer._trainer.init_sigma_method.RANDOM_SIGMA¶

VARIANCE_DATA = bob.trainer._trainer.init_sigma_method.VARIANCE_DATA¶

VARIANCE_G = bob.trainer._trainer.init_sigma_method.VARIANCE_G¶

WITHIN_SCATTER = bob.trainer._trainer.init_g_method.WITHIN_SCATTER¶

e_step((EMTrainerPLDA)self, (PLDABase)machine, (object)data) → None :¶: Updates the hidden variable distribution (or the sufficient statistics) given the Machine parameters.

enrol((PLDATrainer)self, (PLDAMachine)plda_machine, (object)data) → None :¶: Enrol a class-specific model (PLDAMachine) given a set of enrolment samples.

finalize((EMTrainerPLDA)self, (PLDABase)machine, (object)data) → None :¶: This method is called at the end of the EM algorithm

init_f_method¶: The method used for the initialization of F.

init_f_ratio¶: The ratio used for the initialization of F.

init_g_method¶: The method used for the initialization of G.

init_g_ratio¶: The ratio used for the initialization of G.

init_sigma_method¶: The method used for the initialization of sigma.

init_sigma_ratio¶: The ratio used for the initialization of sigma.

initialize((EMTrainerPLDA)self, (PLDABase)machine, (object)data) → None :¶: This method is called before the EM algorithm

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

m_step((EMTrainerPLDA)self, (PLDABase)machine, (object)data) → None :¶: Updates the Machine parameters given the hidden variable distribution (or the sufficient statistics)

max_iterations¶: Max iterations

rng¶: The Mersenne Twister mt19937 random generator used for the initialization of subspaces/arrays before the EM loop.

train((EMTrainerPLDA)self, (PLDABase)machine, (object)data) → None :¶: Trains a PLDABase using data (mu, F, G and sigma are learnt).

use_sum_second_order¶: Tells whether the second order statistics are stored during the training procedure, or only their sum.

z_first_order¶

z_second_order¶

z_second_order_sum¶