bsfit.nodes.models.utils.get_bayes_lookup

bsfit.nodes.models.utils.get_bayes_lookup(percept_space: numpy.array, stim_mean: numpy.array, k_llh: float, prior_mode: float, k_prior: float, prior_shape: str, k_card: float, readout: str)[source]

Create a bayes lookup matrix based on Girshick paper

Parameters

  • percept_space (np.ndarray) – the percept space (1:1:360)

  • stim_mean (np.ndarray) – the stimulus features

  • k_llh (float) – the likelihood concentration

  • prior_mode (float) – the mode of the prior

  • k_prior (float) – the prior concentrations

  • prior_shape (str) – the prior function

  • k_card (float) – the cardinal prior concentration

  • readout (str) – the decision process (“map”)

Usage:
percept, logl_percept = get_bayes_lookup(
    percept_space=np.arange([0,360,1]),
    stim_mean=np.arange([0,360,5]),
    k_llh=5.0,
    prior_mode=225.0,
    k_prior=4.77,
    k_card=0.0,
    prior_tail=0.0,
    prior_shape='vonMisesPrior',
    )
Returns

percepts (np.ndarray): percept likelihood (M_measurements x N_stimulus features)

Return type

(np.ndarray)