Python API for bob.bio.gmm

Todo

Improve documentation of the functions and classes of bob.bio.gmm.

Generic functions

Miscellaneous functions

bob.bio.base.get_config() Returns a string containing the configuration information.

Tools to run recognition experiments

Command line generation

bob.bio.gmm.tools.add_parallel_gmm_options(parsers) Add the options for parallel UBM training to the given parsers.
bob.bio.gmm.tools.initialize_parallel_gmm(args)
bob.bio.gmm.tools.add_jobs(args, submitter, …) Adds all (desired) jobs of the tool chain to the grid, or to the local list to be executed.

Parallel GMM

bob.bio.gmm.tools.kmeans_initialize(…[, …]) Initializes the K-Means training (non-parallel).
bob.bio.gmm.tools.kmeans_estep(algorithm, …) Performs a single E-step of the K-Means algorithm (parallel)
bob.bio.gmm.tools.kmeans_mstep(algorithm, …) Performs a single M-step of the K-Means algorithm (non-parallel)
bob.bio.gmm.tools.gmm_initialize(algorithm, …) Initializes the GMM calculation with the result of the K-Means algorithm (non-parallel).
bob.bio.gmm.tools.gmm_estep(algorithm, …) Performs a single E-step of the GMM training (parallel).
bob.bio.gmm.tools.gmm_mstep(algorithm, …) Performs a single M-step of the GMM training (non-parallel)
bob.bio.gmm.tools.gmm_project(algorithm, …) Performs GMM projection

Parallel ISV

bob.bio.gmm.tools.train_isv(algorithm[, …]) Finally, the UBM is used to train the ISV projector/enroller.

Parallel I-Vector

bob.bio.gmm.tools.ivector_estep(algorithm, …) Performs a single E-step of the IVector algorithm (parallel)
bob.bio.gmm.tools.ivector_mstep(algorithm, …) Performs a single M-step of the IVector algorithm (non-parallel)
bob.bio.gmm.tools.ivector_project(algorithm, …) Performs IVector projection
bob.bio.gmm.tools.train_whitener(algorithm) Train the feature projector with the extracted features of the world group.

Integration with bob.bio.video

bob.bio.gmm.tools.is_video_extension(algorithm)
bob.bio.gmm.tools.base(algorithm) Returns the base algorithm, if it is a video extension, otherwise returns the algorithm itself
bob.bio.gmm.tools.read_feature(extractor, …)

Details

bob.bio.gmm.tools.add_jobs(args, submitter, local_job_adder)[source]

Adds all (desired) jobs of the tool chain to the grid, or to the local list to be executed.

bob.bio.gmm.tools.add_parallel_gmm_options(parsers, sub_module=None)[source]

Add the options for parallel UBM training to the given parsers.

bob.bio.gmm.tools.base(algorithm)[source]

Returns the base algorithm, if it is a video extension, otherwise returns the algorithm itself

bob.bio.gmm.tools.gmm_estep(algorithm, extractor, iteration, indices, force=False, allow_missing_files=False)[source]

Performs a single E-step of the GMM training (parallel).

bob.bio.gmm.tools.gmm_initialize(algorithm, extractor, limit_data=None, force=False, allow_missing_files=False)[source]

Initializes the GMM calculation with the result of the K-Means algorithm (non-parallel). This might require a lot of memory.

bob.bio.gmm.tools.gmm_mstep(algorithm, iteration, number_of_parallel_jobs, force=False, clean=False)[source]

Performs a single M-step of the GMM training (non-parallel)

bob.bio.gmm.tools.gmm_project(algorithm, extractor, indices, force=False, allow_missing_files=False)[source]

Performs GMM projection

bob.bio.gmm.tools.initialize_parallel_gmm(args, sub_module=None)[source]
bob.bio.gmm.tools.is_video_extension(algorithm)[source]
bob.bio.gmm.tools.ivector_estep(algorithm, iteration, indices, force=False, allow_missing_files=False)[source]

Performs a single E-step of the IVector algorithm (parallel)

bob.bio.gmm.tools.ivector_mstep(algorithm, iteration, number_of_parallel_jobs, force=False, clean=False)[source]

Performs a single M-step of the IVector algorithm (non-parallel)

bob.bio.gmm.tools.ivector_project(algorithm, indices, force=False, allow_missing_files=False)[source]

Performs IVector projection

bob.bio.gmm.tools.kmeans_estep(algorithm, extractor, iteration, indices, force=False, allow_missing_files=False)[source]

Performs a single E-step of the K-Means algorithm (parallel)

bob.bio.gmm.tools.kmeans_initialize(algorithm, extractor, limit_data=None, force=False, allow_missing_files=False)[source]

Initializes the K-Means training (non-parallel).

bob.bio.gmm.tools.kmeans_mstep(algorithm, iteration, number_of_parallel_jobs, force=False, clean=False)[source]

Performs a single M-step of the K-Means algorithm (non-parallel)

bob.bio.gmm.tools.lda_project(algorithm, indices, force=False, allow_missing_files=False)[source]

Performs IVector projection

bob.bio.gmm.tools.read_feature(extractor, feature_file)[source]
bob.bio.gmm.tools.save_projector(algorithm, force=False)[source]
bob.bio.gmm.tools.train_isv(algorithm, force=False, allow_missing_files=False)[source]

Finally, the UBM is used to train the ISV projector/enroller.

bob.bio.gmm.tools.train_lda(algorithm, force=False, allow_missing_files=False)[source]

Train the feature projector with the extracted features of the world group.

bob.bio.gmm.tools.train_plda(algorithm, force=False, allow_missing_files=False)[source]

Train the feature projector with the extracted features of the world group.

bob.bio.gmm.tools.train_wccn(algorithm, force=False, allow_missing_files=False)[source]

Train the feature projector with the extracted features of the world group.

bob.bio.gmm.tools.train_whitener(algorithm, force=False, allow_missing_files=False)[source]

Train the feature projector with the extracted features of the world group.

bob.bio.gmm.tools.wccn_project(algorithm, indices, force=False, allow_missing_files=False)[source]

Performs IVector projection

bob.bio.gmm.tools.whitening_project(algorithm, indices, force=False, allow_missing_files=False)[source]

Performs IVector projection