bsfit.nodes.models.utils

Useful functions for modeling

Copyright 2022 by Steeve Laquitaine, GNU license

Functions

choose_percept(readout, stim_mean_space, ...)	choose the percepts for measurement m_i caused by stimulus s_i
do_bayes_inference(k_llh, prior_mode, ...)	do bayesian inference
fit_maxlogl(database, init_p, prior_shape, ...)	Fit stimulus feature estimates
flatten(x)	flattens list of list
format_params(database, init_p, prior_shape, ...)	set model and task parameters
get_bayes_lookup(percept_space, stim_mean, ...)	Create a bayes lookup matrix based on Girshick paper
get_combination_set(database)	get the set of row combinations
get_data(database)	extract the data to fit from the database
get_data_stats(data, output)	calculate data statistics
get_fit_variables(fit_p, params, stim_mean, ...)	get the fit intermediate variable
get_learnt_prior(percept_space, prior_mode, ...)	calculate the learnt prior density
get_logl(fit_p, params, stim_mean, data)	calculate the log(likelihood) of the data given the model
get_logl_and_aic(n_fit_params, proba_data)	calculate - log(likelihood) and akaike information criterion
get_percept_likelihood(percept_space, ...)	calculate percept likelihoods.
get_prediction_stats(output)	calculate prediction statistics
get_proba_data(estimate, proba_estimate)	get data probability density
get_proba_estimate(k_m, PupoGivenModel)	get estimate probability density
get_proba_percept(stim_mean, params, k_llh, ...)	get the percept probability density
get_trial_prediction(output, params, n_repeats)	get model-generated stochastic choices
locate_fit_params(params)	find fit parameters in dictionary
predict(fit_p, params, stim_mean, data, ...)	""get model prediction at trial or summary statistics level
simulate(database, sim_p, prior_shape, ...)	simulate estimate data per condition
simulate_dataset(fit_p, params, stim_mean, ...)	""get model predictions
unpack(my_dict)	unpack a dictionary into a flat list