pyhrf.sandbox.stats.GSVariable(name, initialization, do_sampling=True, axes_names=None, axes_domains=None)¶check_against_truth(atol, rtol, inaccuracy_handling='print')¶check_initialization_arg(ia)¶enable_sampling(flag=True)¶get_accuracy_against_truth(abs_error, rel_error, fv, tv, atol, rtol)¶Return the accuray of the estimate fv, compared to the true value tv
get_custom_init()¶Must return a numpy.ndarray. Consider initializing with a good guess so that sampling converges more quickly.
get_estim_value_for_check()¶get_random_init()¶Must return a random numpy.ndarray that will then be used as init value for sampling. For example, it can be a sample from the prior distribution. This function will also be used to test for the sensitivity to initialization.
get_true_value_for_check()¶get_variable(vname)¶get_variable_value(vname)¶Short-hand to get variable among all those defined in the parent sampler
init_observables()¶init_sampling()¶reset()¶sample()¶Draw a sample conditionally to the current Gibbs Sampler state. Must return a numpy.ndarray.
Variables which have been registered in the parent GibbsSampler object can be retrieved via methods self.get_variable(var_name) and self.get_variable_value(var_name)
set_init_value()¶Set the initial value of self.current_value, depending on the initialization scenario (random, custom, truth).
set_initialization(init)¶set_outputs(outputs, output_type='ndarray')¶| Parameters: |
|
|---|
Return: None
set_true_value(true_value)¶track_obs_quantity(name, quantity, history_pace=None, axes_names=None, axes_domains=None)¶track_sampled_quantity(name, quantity, history_pace=None, axes_names=None, axes_domains=None)¶update_observables()¶Update quantities after the burnin period
pyhrf.sandbox.stats.GibbsSampler(sampled_variables, nb_its_max, obs_pace=1, burnin=0.3, sample_hist_pace=-1, obs_hist_pace=-1)¶check_against_truth(default_atol=0.1, default_rtol=0.1, var_specific_atol=None, var_specific_rtol=None, inaccuracy_handling='print')¶get_outputs(output_type='ndarray')¶output_type : ‘ndarray’ or ‘cuboid’
get_variable(vname)¶get_variable_value(vname)¶iterate_sampling()¶reset()¶run()¶set_initialization(vname, init)¶set_true_value(vname, true_value)¶set_true_values(true_values)¶set_variable(name, var)¶set_variables(var_dict)¶stop_criterion(iteration)¶track_obs_quantity(name, q, history_pace=None, axes_names=None, axes_domains=None)¶track_sampled_quantity(name, q, history_pace=None, axes_names=None, axes_domains=None)¶pyhrf.sandbox.stats.Trajectory(variable, history_pace, history_start, max_iterations, init_iteration=None, axes_names=None, axes_domains=None)¶Keep track of a numpy array that is modified _inplace_ iteratively TODO: when mature, should be moved to pyhrf.ndarray should replace pyhrf.jde.samplerbase.Trajectory
get_last()¶Return the last saved element
to_cuboid()¶Pack the current trajectory in a xndarray
update(iteration)¶Record the current variable value