KJB
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
Class List
Here are the classes, structs, unions and interfaces with brief descriptions:
[detail level 1234]
oNboost
oNDTLib
oNergo
oNkjbClasses and functions for dealing with trajectory files
oNkjb_c
oNkjb_parallel
oNQuaternion
oNsemantics
oNspear
oNstd
oCa
oCAbstractAbstract class to render this object
oCAbstract_gibbs_step
oCAbstract_hmc_step
oCAbstract_mh_step
oCAbstract_sampler
oCAll_log_recorder
oCAll_model_recorder
oCAnnealable
oCAnnealing_mh_step
oCAnnealing_proposer_wrapper
oCAnnealing_sampler
oCbase_model_archetype
oCBaseModel
oCBasic_gibbs_step
oCBasic_hmc_step
oCBasic_mh_step
oCBasic_sd_step
oCBest_model_recorder
oCBest_target_recorder
oCBlobA simple class that represents a blob
oCBlob_detectorA blob detector class. Use operator() to apply to image
oCCallback_recorder
oCCompute_blob
oCComputes
oCConditional_distribution_proposer
oCConstant_parameter_evaluatorReturns the same result no matter what model is received
oCConstrained_targetAdapts a target distribution to be one with bounds
oCControl_sceneClass that represents a scene plus the control ouptuts; i.e. the model for the control-point trajectory sampler
oCCurrent_log_recorder
oCCurrent_model_recorder
oCEvent_listener
oCExpectation_recorder
oCFace_data
oCFeaturesAllows to manipulate basic 2d image features (edges, fitted line segments and manhattan worl). This is mostly useful for Manhattan world scenes so it might be renamed. This class contains the following features:
oCGaussian_random_walk_proposer
oCGaussian_scale_space
oCGaussian_scale_space_generator
oCGeneric_adaptive_mh_step
oCGet_model_parameterGets the specified parameter of a model. For now, we assume all parameters are type double
oCGibbs_model_proposer
oCGLUT_polymesh
oCgsl_matrix
oCgsl_multifit_function_fdf
oCgsl_multimin_function
oCgsl_multimin_function_fdf
oCGsl_rng_cmrgRandom number generator using L'Ecuyer's 1996 algorithm. This implements the Combined Multiple Recursive Generator algorithm of L'Ecuyer (1996). Period is 10 ** 56. Defined using macro Gsl_rng_template
oCGsl_rng_gfsr4Random number generator using a four-tap XOR using a shift register. This uses Ziff's offsets (1998) and is very fast
oCGsl_rng_mrgRandom number generator using 1993 algorithm of L'Ecuyer et al. This implements the Multiple Recursive Generator algorithm of L'Ecuyer et al. (1993). Period is 10 ** 46. Defined using macro Gsl_rng_template
oCGsl_rng_mt19937Random number generator using the "Mersenne Twister" algorithm. This implements the "Mersenne Twister" of Matsumoto and Nishimura. The period is about 10 ** 6000. This is an all-around good PRNG. Defined using macro Gsl_rng_template
oCGsl_rng_ranlxd1Random number generator using the "RANLUX" algorithm, 48 bits, lvl. 1 This implements the "RANLUX" algorithm of Luescher at double precision. This is a "luxury random number" algorithm, i.e., slow. This one is "level 1" so it's not as decorrelated as level 2. Period is 10 ** 171. Defined using macro Gsl_rng_template
oCGsl_rng_ranlxd2Random number generator using the "RANLUX" algorithm, 48 bits, lvl. 1 This implements the "RANLUX" algorithm of Luescher at double precision. This is a "luxury random number" algorithm, i.e., slow. This one is "level 2" so it's the most decorrelated. Period is 10 ** 171. Defined using macro Gsl_rng_template
oCGsl_rng_ranlxs0Random number generator using the "RANLUX" algorithm, 24 bits. This implements the "RANLUX" algorithm of Luescher at single precision, i.e., meant for a float not a double. This is a "luxury random number" algorithm, i.e., slow. Nevertheless this one is "level 0" so it's the entry-level luxury model. Period is 10 ** 171. Defined using macro Gsl_rng_template
oCGsl_rng_ranlxs1Random number generator using the "RANLUX" algorithm, 24 bits. This implements the "RANLUX" algorithm of Luescher at single precision, i.e., meant for a float not a double. This is a "luxury random number" algorithm, i.e., slow. This one is "level 1" so it's the mid-level luxury model. Period is 10 ** 171. Defined using macro Gsl_rng_template
oCGsl_rng_ranlxs2Random number generator using the "RANLUX" algorithm, 24 bits. This implements the "RANLUX" algorithm of Luescher at single precision, i.e., meant for a float not a double. This is a "luxury random number" algorithm, i.e., slow. This one is "level 2" so it's the top-level luxury model. Period is 10 ** 171. Defined using macro Gsl_rng_template
oCGsl_rng_taus2Random number generator using Tausworthe's algorithm. This is L'Ecuyer's version of Tausworthe's algorithm (or something like that). Period is 10 ** 26. Defined using macro Gsl_rng_template
oCgsl_vector
oCIndependent_gaussian_proposer
oCLine
oCMh_model_proposer
oCMh_proposal_resultIndicates the result of an MH proposal. It is simply a pair of probabilities, forward and backward
oCModel_dimensionReturns the dimension of the model
oCModel_edge
oCModel_evaluator
oCmodel_evaluator_archetype
oCModel_parameter_evaluator
oCmodel_proposer_archetype
oCModel_recorder
oCmodel_recorder_archetype
oCModelEvaluator
oCModelProposer
oCModelRecorder
oCModulo_recorder
oCMove_model_parameterMoves the specified parameter of a model. For now, we assume all parameters are type double
oCMove_model_parameter_as_plusDefault move function; uses '+'
oCMulti_model_recorder
oCMulti_proposer_proposer
oCMulti_step_sampler
oCMv_gaussian_proposer
oCNull_event_listener
oCNull_recorder
oCNull_value
oCNumerical_gradientApproximates the gradient and/or curvature of a target distribution, evaluated at a certain location. The user must provide the mechanisms to change the model (see constructor)
oCOstream_recorder
oCPosteriorGeneric posterior class
oCReal_hmc_step
oCReal_numerical_gradientApproximates the gradient of a target distribution, evaluated at a certain location. The model in question must be a vector model
oCReal_sd_step
oCRecent_log_recorder
oCRecent_model_recorder
oCSampler_step
oCSerialize_recorder
oCSet_model_parameterSets the specified parameter of a model. For now, we assume all parameters are type double
oCSimple_adaptive_mh_step
oCSingle_dimension_proposer
oCSingle_step_sampler
oCST_SPHERE
oCStep_log
oCStep_resultStructure for returning results of a single sampling move
oCtriangulateio
oCUpdatable
oCVector_hmc_step
oCvector_model_archetype
oCVector_numerical_gradientApproximates the gradient of a target distribution, evaluated at a certain location. The model in question must be a vector model
oCVector_sd_step
oCVector_srw_step
oCVectorModel
\CViewing_recorder