def test_participant_processing3(self, participant_data_setup):
participant_data = participant_data_setup
fit_sim = fitSims.FitSim()
result = fit_sim.participant_sequence_generation(
participant_data, 'Decisions', 'Rewards', None,
['A', 'B', 'C', 'D', 'E', 'F'])
correct_result = [[
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F']),
(None, ['A', 'B', 'C', 'D', 'E', 'F'])
], ['E', 'F', 'B', 'E', 'E', 'F', 'D', 'E', 'C', 'C', 'A', 'A'],
[
1.0, 0.0, 0.0, 1.0, 0.0, 1.0, np.nan, np.nan,
np.nan, np.nan, np.nan, np.nan
]]
for row, correct_row in zip(result, correct_result):
assert row == correct_row
def test_FS_subset2(self):
fit_sim = fitSims.FitSim()
subset_values = ['boo', 1, {}]
for values in subset_values:
with pytest.raises(
fitSims.FitSubsetError,
match='{} is not a known fit_subset'.format(values)):
assert fit_sim._preprocess_fit_subset(values)
def test_FS_subset(self):
fit_sim = fitSims.FitSim()
subset_values = [
np.nan, 'all', 'unrewarded', 'rewarded', [1, 2, 3], None
]
subset_returns = [[], None, [], [], [1, 2, 3], None]
for values, returns in zip(subset_values, subset_returns):
results = fit_sim._preprocess_fit_subset(values)
assert results == returns
def test_participant_processing4(self, participant_data_setup):
participant_data = participant_data_setup
fit_sim = fitSims.FitSim()
with pytest.raises(fitSims.ActionError):
fit_sim.participant_sequence_generation(participant_data,
'Decisions', 'Rewards',
None,
['A', 'C', 'D', 'E', 'F'])
def test_FS_subset4(self):
fit_sim = fitSims.FitSim()
part_rewards = [1, 2, np.nan, np.nan, 5]
subset_values = ['all', None]
for values in subset_values:
with pytest.raises(
fitSims.FitSubsetError,
match='{} is not a known fit_subset'.format(values)):
assert fit_sim._set_fit_subset(values, part_rewards)
def test_model_parameters(self, model_setup):
model, (model_parameters, model_properties) = model_setup
fit_sim = fitSims.FitSim()
fit_sim.model_parameter_names = list(model_parameters.keys())
results = fit_sim.get_model_parameters(
*list(model_parameters.values()))
correct_results = model_parameters
for k in results:
assert results[k] == correct_results[k]
def test_FS_subset3(self):
fit_sim = fitSims.FitSim()
part_rewards = [1, 2, np.nan, np.nan, 5]
subset_values = [np.nan, 'unrewarded', 'rewarded']
subset_returns = [[False, False, True, True, False],
[False, False, True, True, False],
[True, True, False, False, True]]
for values, returns in zip(subset_values, subset_returns):
results = fit_sim._set_fit_subset(values, part_rewards)
assert all(results == returns)
def test_FS_info(self):
fit_sim = fitSims.FitSim()
result = fit_sim.info()
correct_result = {
'Name': 'FitSim',
'participant_choice_property': 'Actions',
'participant_reward_property': 'Rewards',
'task_stimuli_property': None,
'action_options_property': None,
'model_fitting_variable': 'ActionProb',
'float_error_response_value': 1 / 1e100,
'fit_subset': None
}
for k in result.keys():
assert result[k] == correct_result[k]
def test_prepare_sim(self, model_setup, participant_data_setup):
model, model_other = model_setup
participant_data = participant_data_setup
fit_sim = fitSims.FitSim(
participant_choice_property='Decisions',
participant_reward_property='Rewards',
model_fitting_variable='ActionProb',
task_stimuli_property=None,
fit_subset=None,
action_options_property='valid_actions_combined')
fitting = fit_sim.prepare_sim(model, model_other[0], model_other[1],
participant_data)
assert fitting == fit_sim.fitness
def test_model_properties(self, model_setup):
model, (model_parameters, model_properties) = model_setup
fit_sim = fitSims.FitSim()
fit_sim.model_parameter_names = list(model_parameters.keys())
fit_sim.model_other_properties = model_properties
results = fit_sim.get_model_properties(
*list(model_parameters.values()))
correct_results = model_parameters.copy()
correct_results.update(model_properties)
for k in results:
result = results[k]
if isinstance(result, np.ndarray):
assert all(result == correct_results[k])
else:
assert result == correct_results[k]
def test_fitness(self, model_setup, participant_data_setup):
model, model_other = model_setup
participant_data = participant_data_setup
fit_sim = fitSims.FitSim(participant_choice_property='Choices',
participant_reward_property='Rewards',
model_fitting_variable='ActionProb',
task_stimuli_property=None,
fit_subset=None,
action_options_property=[1, 2])
fitness = fit_sim.prepare_sim(model, model_other[0], model_other[1],
participant_data)
result = fitness(0.5, 3)
correct_result = np.array([
0.5, 0.64565631, 0.46257015, 0.62831619, 0.55601389, 0.51874122,
0.5093739, 0.4906261, 0.4906261, 0.5093739, 0.5093739, 0.5093739
])
np.testing.assert_array_almost_equal(result, correct_result)
def test_FS_basic(self):
fit_sim = fitSims.FitSim()
assert isinstance(fit_sim, fitSims.FitSim)
