Fit

Tools for fitting models to free-recall data.

The fit module provides tools for fitting a model to free-recall data. Models can inherit from the Recall class and implement subject-level methods for evaluating likelihood and simulating data. The model class will then have access to high-level methods for evaluating, fitting, and simulating models.

Utilities

prepare_lists(data[, study_keys, ...])	Prepare study and recall data for simulation.
prepare_study(study_data[, study_keys])	Prepare study phase data for simulation.
get_best_results(results)	Get best results from a repeated search.
add_recalls(study, recalls_list)	Add recall sequences to a study DataFrame.

Model evaluation

Recall()	Base class for evaluating a model of free recall.
Recall.likelihood_subject(study, recall, param)	Log likelihood of data for one subject based on a given model.
Recall.likelihood(data, group_param[, ...])	Log likelihood summed over all subjects.

Parameter estimation

Recall.fit_subject(subject_data, param_def)	Fit a model to data for one subject.
Recall.fit_indiv(data, param_def[, ...])	Fit parameters to individual subjects.

Generating simulated data

Recall.generate_subject(study, recall, param)	Generate simulated data for one subject.
Recall.generate(data, group_param[, ...])	Generate simulated data for all subjects.

Characterizing model parameters

Recall.parameter_sweep(data, group_param, ...)	Simulate data with varying parameters.
Recall.parameter_recovery(data, n_sample, ...)	Run multiple iterations of parameter recovery.