magni.cs.phase_transition._config moduleΒΆ
Module providing configuration options for the phase_transition subpackage.
See also
magni.utils.config.Configger- The Configger class used
Notes
This module instantiates the Configger class provided by
magni.utils.config. The configuration options are the following:
- algorithm_kwargs : dict
- The keyword arguments passed on to the reconstruction algorithm.
- coefficients : {‘rademacher’, ‘gaussian’, ‘laplace’, ‘bernoulli’}
- The distribution which the non-zero coefficients in the coefficient vector are drawn from.
- custom_noise_factory : callable
- The callable that generates a custom m-by-1 vector of noise from the four arguments “m”, “n”, “k”, “noise_power”.
- custom_system_matrix_factory : callable
- The callable that generates a custom m-by-n system matrix from the two arguments “m”, “n”.
- delta : list or tuple
- The delta values of the delta-rho grid whose points are used for the monte carlo simulations (the default is [0., 1.]).
- logit_solver : {‘built-in’, ‘sklearn’}
- The solver to use in the logistic regression fit of the phase transition.
- maxpoints : int
- The maximum number of phase space grid points to be handled by a process before it is replaced by a new process to free ressources (the default is None, which implies that processes are not replaced).
- monte_carlo : int
- The number of monte carlo simulations to run in each point of the delta-rho grid (the default is 1).
- noise : {‘AWGN’, ‘AWLN’, ‘custom’}
- The type of noise to use (the default is None, which implies that noiseless measurements are used).
- problem_size : int
- The length of the coefficient vector (the default is 800).
- rho : list or tuple
- The rho values of the delta-rho grid whose points are used for the monte carlo simulations (the default is [0., 1.]).
- seed : int
- The seed used when picking seeds for generating data for the monte carlo simulations or using the scikit-learn logistic regression solver (the default is None, which implies an arbitrary seed).
- SNR : int or float
- The signal-to-noise ratio in decibel to use (the default is 40 dB).
- support_distribution : ndarray
- The n x 1 support distribution array to use in the generation of test vectors (the default is None, which implies that the active entries in test vectors are the low index entries).
- system_matrix : {‘USE’, ‘RandomDCT2D’, ‘custom’}
- The system matrix to use in the simulation (the default is USE, which implies that the system matrix is drawn from the Uniform Spherical Ensemble ).