def test_add_s2chain(self):
rnd = np.random.random_sample(size=(3, 2))
key = 's2chain'
RS = ResultsStructure()
RS.add_s2chain(s2chain=rnd)
self.assertTrue(np.array_equal(RS.results[key], rnd),
msg=str('Expect equal arrays: {} != {}'.format(
RS.results[key], rnd)))
def test_add_rndnumseq(self):
rnd = np.random.random_sample(size=(3, 2))
key = 'rndseq'
RS = ResultsStructure()
RS.add_random_number_sequence(rndseq=rnd)
self.assertTrue(np.array_equal(RS.results[key], rnd),
msg=str('Expect equal arrays: {} != {}'.format(
RS.results[key], rnd)))
def test_add_time_stats(self):
key = 'time [mh, dr, am]'
mtime = 0.01
drtime = 0.10
adtime = 0.05
RS = ResultsStructure()
RS.add_time_stats(mtime=mtime, drtime=drtime, adtime=adtime)
self.assertEqual(RS.results[key], [mtime, drtime, adtime],
msg='Lists should be equal.')
def test_resfilename_is_not_none(self):
__, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
RS = ResultsStructure()
RS.add_options(options=options)
RS.results['simulation_options']['results_filename'] = 'test'
filename = RS.determine_filename(
options=RS.results['simulation_options'])
self.assertEqual(filename, 'test', msg='Filename matches')
def test_key_removal(self):
options = SimulationOptions()
options.define_simulation_options(doram=3, nsimu=500)
RS = ResultsStructure()
RS.add_options(options=options)
self.assertEqual(RS.results['simulation_options']['nsimu'],
options.nsimu,
msg=str('nsimu = {}'.format(options.nsimu)))
self.assertFalse('doram' in RS.results['simulation_options'].items(),
msg='doram option should not be saved')
def test_addbasic_false(self):
model, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
DR = DelayedRejection()
DR._initialize_dr_metrics(options)
RS = ResultsStructure()
self.assertFalse(RS.add_dram(drscale=options.drscale,
RDR=covariance._RDR,
total_rejected=rejected['total'],
drsettings=DR),
msg='basic features not added to result structure')
def test_add_prior(self):
RS = ResultsStructure()
mu = 0.0
sigma = 1.0
priortype = 1
RS.add_prior(mu=mu, sigma=sigma, priortype=priortype)
self.assertEqual(RS.results['prior'],
dict(mu=mu, sigma=sigma, priortype=priortype),
msg=str('{} neq {}'.format(
RS.results['prior'],
dict(mu=mu, sigma=sigma, priortype=priortype))))
def test_key_removal(self):
model = ModelSettings()
model.define_model_settings(N=100, sos_function='hello world')
RS = ResultsStructure()
RS.add_model(model=model)
self.assertEqual(RS.results['model_settings']['N'],
model.N,
msg=str('N = {}'.format(model.N)))
self.assertFalse('sos_function'
in RS.results['model_settings'].items(),
msg='sos_function setting should not be saved')
def test_dump_to_json_file(self):
RS = ResultsStructure()
tmpfile = gf.generate_temp_file(extension='json')
results = {'model': 4, 'data': np.random.random_sample(size=(1000, 2))}
RS.save_json_object(results=results, filename=tmpfile)
retres = RS.load_json_object(filename=tmpfile)
self.assertEqual(retres['model'],
results['model'],
msg='Model set to 4')
self.assertTrue(np.array_equal(retres['data'], results['data']),
msg='Arrays should be equal')
os.remove(tmpfile)
def test_resfilename_is_none(self):
__, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
RS = ResultsStructure()
RS.add_options(options=options)
RS.results['simulation_options']['results_filename'] = None
filename = RS.determine_filename(
options=RS.results['simulation_options'])
self.assertEqual(filename,
str('{}{}{}'.format(
RS.results['simulation_options']['datestr'], '_',
'mcmc_simulation.json')),
msg='Filename matches')
def test_updatesigma_false(self):
updatesigma = False
sigma2 = np.ones([2, 1])
S20 = np.ones([2, 1])
N0 = np.ones([2, 1])
RS = ResultsStructure()
RS.add_updatesigma(updatesigma=updatesigma,
sigma2=sigma2,
S20=S20,
N0=N0)
self.assertTrue(np.array_equal(RS.results['sigma2'], sigma2),
msg=str('sigma2 -> {}'.format(sigma2)))
self.assertTrue(np.isnan(RS.results['S20']), msg='S20 -> np.nan')
self.assertTrue(np.isnan(RS.results['N0']), msg='N0 -> np.nan')
def test_load_ser_sim_res(self):
tmpfolder = gf.generate_temp_folder()
tmpfolder0 = os.path.join(tmpfolder, 'chain_0')
os.makedirs(tmpfolder0)
tmpfile0 = 'chain_0.json'
tmpfile0 = os.path.join(tmpfolder0, tmpfile0)
results = dict(a=[0, 1], b='hello')
RS = ResultsStructure()
RS.save_json_object(results, tmpfile0)
pres = CP.load_serial_simulation_results(tmpfolder0)
shutil.rmtree(tmpfolder)
self.assertTrue(isinstance(pres['a'], np.ndarray))
self.assertTrue(isinstance(pres['b'], str))
self.assertEqual(pres['b'], 'hello')
def test_addbasic_true(self):
model, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
RS = ResultsStructure()
RS.add_basic(nsimu=options.nsimu,
covariance=covariance,
parameters=parameters,
rejected=rejected,
simutime=0.001,
theta=chain[-1, :])
self.assertTrue(RS.basic,
msg='basic features added to result structure')
self.assertTrue(np.array_equal(RS.results['theta'], np.array([0, 0])),
msg='Last elements of chain are zero')
def test_addbasic_rejection(self):
model, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
RS = ResultsStructure()
RS.add_basic(nsimu=options.nsimu,
covariance=covariance,
parameters=parameters,
rejected=rejected,
simutime=0.001,
theta=chain[-1, :])
self.assertEqual(RS.results['total_rejected'],
10 * (options.nsimu**(-1)),
msg='rejection reported as fraction of nsimu')
self.assertEqual(RS.results['rejected_outside_bounds'],
2 * (options.nsimu**(-1)),
msg='rejection reported as fraction of nsimu')
def test_addbasic_covariance(self):
model, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
covariance._R[0, 0] = 1.1
covariance._R[0, 1] = 2.3
RS = ResultsStructure()
RS.add_basic(nsimu=options.nsimu,
covariance=covariance,
parameters=parameters,
rejected=rejected,
simutime=0.001,
theta=chain[-1, :])
self.assertTrue(np.array_equal(RS.results['R'], covariance._R),
msg='Cholesky matches')
self.assertTrue(np.array_equal(
RS.results['qcov'], np.dot(covariance._R.transpose(),
covariance._R)),
msg='Covariance matches')
def test_allnames(self):
model, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
RS = ResultsStructure()
RS.add_basic(nsimu=options.nsimu,
covariance=covariance,
parameters=parameters,
rejected=rejected,
simutime=0.001,
theta=chain[-1, :])
allnames = RS.results['allnames']
names = RS.results['names']
print(allnames)
print(names)
self.assertEqual(len(allnames),
len(names) + 1,
msg='Expect extra name in allnames')
self.assertEqual(allnames[-1],
'b2',
msg='Expect final parameter is b2')
def test_addbasic_true(self):
model, options, parameters, data, covariance, rejected, chain, s2chain, sschain = gf.setup_mcmc_case_dr(
)
covariance._RDR = np.random.random_sample(size=(2, 2))
DR = DelayedRejection()
DR._initialize_dr_metrics(options)
DR.dr_step_counter = 12000
RS = ResultsStructure()
RS.add_basic(nsimu=options.nsimu,
covariance=covariance,
parameters=parameters,
rejected=rejected,
simutime=0.001,
theta=chain[-1, :])
self.assertTrue(RS.add_dram(drscale=options.drscale,
RDR=covariance._RDR,
total_rejected=rejected['total'],
drsettings=DR),
msg='basic features added to result structure')
self.assertTrue(np.array_equal(RS.results['RDR'], covariance._RDR),
msg='RDR matches')
self.assertEqual(RS.results['alpha_count'],
DR.dr_step_counter,
msg='Alpha count matches dr step counter')
def test_addbasic_false(self):
RS = ResultsStructure()
self.assertFalse(RS.basic,
msg='basic features not added to result structure')