def test_thermochemical_error_with_multiple_T_X_points(datasets_db):
"""Multiple temperature and composition datapoints in a dataset for a mixing phase should be successful."""
datasets_db.insert(CU_MG_SM_MIX_T_X_FCC_A1)
dbf = Database(CU_MG_TDB)
error = calculate_thermochemical_error(dbf, ['CU', 'MG', 'VA'], list(dbf.phases.keys()), datasets_db)
assert np.isclose(float(error), -31830.303030303032, atol=0.01)
def test_thermochemical_error_with_multiple_X_points(datasets_db):
"""Multiple composition datapoints in a dataset for a mixing phase should be successful."""
datasets_db.insert(CU_MG_CPM_MIX_X_HCP_A3)
dbf = Database(CU_MG_TDB)
error = calculate_thermochemical_error(dbf, ['CU','MG','VA'], list(dbf.phases.keys()), datasets_db)
assert np.isclose(float(error), -520.0, atol=0.01)
def test_thermochemical_error_with_multiple_T_points(datasets_db):
"""Multiple temperature datapoints in a dataset for a stoichiometric comnpound should be successful."""
datasets_db.insert(CU_MG_HM_MIX_T_CUMG2)
dbf = Database(CU_MG_TDB)
error = calculate_thermochemical_error(dbf, ['CU','MG','VA'], list(dbf.phases.keys()), datasets_db)
assert np.isclose(float(error), -3672.727272727273, atol=0.01)
def test_thermochemical_error_for_mixing_entropy_error_is_excess_only(
datasets_db):
"""Tests that error in mixing entropy data is excess only (the ideal part is removed)."""
# If this fails, make sure the ideal mixing contribution is removed.
phase_models = {
"components": ["AL", "B"],
"phases": {
"LIQUID": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
},
"FCC_A1": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
}
}
}
dataset_excess_mixing = {
"components": ["AL", "B"],
"phases": ["FCC_A1"],
"solver": {
"sublattice_site_ratios": [1],
"sublattice_occupancies": [[[0.5, 0.5]]],
"sublattice_configurations": [[["AL", "B"]]],
"mode": "manual"
},
"conditions": {
"P": 101325,
"T": 298.15
},
"output": "SM_MIX",
"values": [[[10]]],
"excluded_model_contributions": ["idmix"]
}
datasets_db.insert(dataset_excess_mixing)
dbf = generate_parameters(phase_models, datasets_db, 'SGTE91', 'linear')
assert dbf.elements == {'AL', 'B'}
assert set(dbf.phases.keys()) == {'LIQUID', 'FCC_A1'}
assert len(dbf._parameters.search(where('parameter_type') == 'L')) == 1
phases = list(dbf.phases.keys())
comps = list(dbf.elements)
# the error should be exactly 0 because we are only fitting to one point
# the dataset is excess only
zero_error_prob = scipy.stats.norm(loc=0, scale=0.2).logpdf(
0.0) # SM weight = 0.2
# Explicitly pass parameters={} to not try fitting anything
thermochemical_data = get_thermochemical_data(dbf,
comps,
phases,
datasets_db,
symbols_to_fit=[])
error = calculate_thermochemical_error(dbf, comps, thermochemical_data)
assert np.isclose(error, zero_error_prob, atol=1e-6)
def lnprob(
params,
comps=None,
dbf=None,
phases=None,
datasets=None,
symbols_to_fit=None,
phase_models=None,
scheduler=None,
callables=None,
thermochemical_callables=None,
):
"""
Returns the error from multiphase fitting as a log probability.
"""
starttime = time.time()
parameters = {
param_name: param
for param_name, param in zip(symbols_to_fit, params)
}
try:
multi_phase_error = calculate_zpf_error(dbf,
comps,
phases,
datasets,
phase_models,
parameters=parameters,
callables=callables)
except (ValueError, LinAlgError) as e:
multi_phase_error = [np.inf]
multi_phase_error = [
np.inf if np.isnan(x) else x**2 for x in multi_phase_error
]
multi_phase_error = -np.sum(multi_phase_error)
single_phase_error = calculate_thermochemical_error(
dbf,
comps,
phases,
datasets,
parameters,
phase_models=phase_models,
callables=thermochemical_callables)
actvity_error = calculate_activity_error(dbf,
comps,
phases,
datasets,
parameters=parameters,
phase_models=phase_models,
callables=callables)
total_error = multi_phase_error + single_phase_error + actvity_error
logging.debug(
'Single phase error: {:0.2f}. Multi phase error: {:0.2f}. Activity Error: {:0.2f}. Total error: {:0.2f}'
.format(single_phase_error, multi_phase_error, actvity_error,
total_error))
logging.debug('lnprob time: {}'.format(time.time() - starttime))
return np.array(total_error, dtype=np.float64)
def test_datasets_convert_thermochemical_string_values_producing_correct_value(
datasets_db):
"""Strings where floats are expected should give correct answers for thermochemical datasets"""
datasets_db.insert(
clean_dataset(CU_MG_DATASET_THERMOCHEMICAL_STRING_VALUES))
dbf = Database(CU_MG_TDB)
error = calculate_thermochemical_error(dbf, ['CU', 'MG', 'VA'],
list(dbf.phases.keys()),
datasets_db)
assert np.isclose(float(error), -3672.727272727273, atol=0.01)
def test_thermochemical_error_with_multiple_T_X_points(datasets_db):
"""Multiple temperature and composition datapoints in a dataset for a mixing phase should be successful."""
datasets_db.insert(CU_MG_SM_MIX_T_X_FCC_A1)
dbf = Database(CU_MG_TDB)
phases = list(dbf.phases.keys())
comps = ['CU', 'MG', 'VA']
thermochemical_data = get_thermochemical_data(dbf, comps, phases,
datasets_db)
error = calculate_thermochemical_error(dbf, comps, thermochemical_data)
assert np.isclose(float(error), -3282497.2380024833, rtol=1e-6)
def test_thermochemical_error_with_multiple_T_points(datasets_db):
"""Multiple temperature datapoints in a dataset for a stoichiometric comnpound should be successful."""
datasets_db.insert(CU_MG_HM_MIX_T_CUMG2)
dbf = Database(CU_MG_TDB)
phases = list(dbf.phases.keys())
comps = ['CU', 'MG', 'VA']
thermochemical_data = get_thermochemical_data(dbf, comps, phases,
datasets_db)
error = calculate_thermochemical_error(dbf, comps, thermochemical_data)
assert np.isclose(error, -14.287293263253728, rtol=1e-6)
def test_mixing_energies_are_fit(datasets_db):
"""Tests that given mixing energy data, the excess parameter is fit."""
phase_models = {
"components": ["AL", "B"],
"phases": {
"LIQUID": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
},
"FCC_A1": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
}
}
}
dataset_excess_mixing = {
"components": ["AL", "B"],
"phases": ["FCC_A1"],
"solver": {
"sublattice_site_ratios": [1],
"sublattice_occupancies": [[[0.5, 0.5]]],
"sublattice_configurations": [[["AL", "B"]]],
"mode": "manual"
},
"conditions": {
"P": 101325,
"T": 298.15
},
"output": "HM_MIX",
"values": [[[-10000]]]
}
datasets_db.insert(dataset_excess_mixing)
dbf = generate_parameters(phase_models, datasets_db, 'SGTE91', 'linear')
assert dbf.elements == {'AL', 'B'}
assert set(dbf.phases.keys()) == {'LIQUID', 'FCC_A1'}
assert len(dbf._parameters.search(where('parameter_type') == 'L')) == 1
assert dbf.symbols['VV0000'] == -40000
# check that read/write is ok
read_dbf = dbf.from_string(dbf.to_string(fmt='tdb'), fmt='tdb')
assert read_dbf.elements == {'AL', 'B'}
assert set(read_dbf.phases.keys()) == {'LIQUID', 'FCC_A1'}
assert len(
read_dbf._parameters.search(where('parameter_type') == 'L')) == 1
# the error should be exactly 0 because we are only fitting to one point
from espei.error_functions import calculate_thermochemical_error
assert calculate_thermochemical_error(read_dbf, sorted(dbf.elements),
sorted(dbf.phases.keys()),
datasets_db) == 0
def test_thermochemical_error_with_multiple_X_points(datasets_db):
"""Multiple composition datapoints in a dataset for a mixing phase should be successful."""
datasets_db.insert(CU_MG_CPM_MIX_X_HCP_A3)
dbf = Database(CU_MG_TDB)
phases = list(dbf.phases.keys())
comps = ['CU', 'MG', 'VA']
thermochemical_data = get_thermochemical_data(dbf, comps, phases,
datasets_db)
error = calculate_thermochemical_error(dbf, comps, thermochemical_data)
assert np.isclose(error, -4061.119001241541, rtol=1e-6)
def test_thermochemical_error_for_mixing_entropy_error_is_excess_only(
datasets_db):
"""Tests that error in mixing entropy data is excess only (the ideal part is removed)."""
# If this fails, make sure the ideal mixing contribution is removed.
phase_models = {
"components": ["AL", "B"],
"phases": {
"LIQUID": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
},
"FCC_A1": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
}
}
}
dataset_excess_mixing = {
"components": ["AL", "B"],
"phases": ["FCC_A1"],
"solver": {
"sublattice_site_ratios": [1],
"sublattice_occupancies": [[[0.5, 0.5]]],
"sublattice_configurations": [[["AL", "B"]]],
"mode": "manual"
},
"conditions": {
"P": 101325,
"T": 298.15
},
"output": "SM_MIX",
"values": [[[10]]]
}
datasets_db.insert(dataset_excess_mixing)
dbf = generate_parameters(phase_models, datasets_db, 'SGTE91', 'linear')
assert dbf.elements == {'AL', 'B'}
assert set(dbf.phases.keys()) == {'LIQUID', 'FCC_A1'}
assert len(dbf._parameters.search(where('parameter_type') == 'L')) == 1
# the error should be exactly 0 because we are only fitting to one point
# the dataset is excess only
from espei.error_functions import calculate_thermochemical_error
assert calculate_thermochemical_error(dbf, sorted(dbf.elements),
sorted(dbf.phases.keys()),
datasets_db) == 0
def predict(params, **ctx):
"""
Calculate lnprob = lnlike + lnprior
"""
logging.debug('Parameters - {}'.format(params))
# lnprior
prior_rvs = ctx['prior_rvs']
lnprior_multivariate = [rv.logpdf(theta) for rv, theta in zip(prior_rvs, params)]
logging.debug('Priors: {}'.format(lnprior_multivariate))
lnprior = np.sum(lnprior_multivariate)
if np.isneginf(lnprior):
# It doesn't matter what the likelihood is. We can skip calculating it to save time.
logging.log(TRACE, 'Proposal - lnprior: {:0.4f}, lnlike: {}, lnprob: {:0.4f}'.format(lnprior, np.nan, lnprior))
return lnprior
# lnlike
parameters = {param_name: param for param_name, param in zip(ctx['symbols_to_fit'], params)}
zpf_kwargs = ctx.get('zpf_kwargs')
activity_kwargs = ctx.get('activity_kwargs')
thermochemical_kwargs = ctx.get('thermochemical_kwargs')
starttime = time.time()
if zpf_kwargs is not None:
try:
multi_phase_error = calculate_zpf_error(parameters=parameters, **zpf_kwargs)
except (ValueError, np.linalg.LinAlgError) as e:
raise e
print(e)
multi_phase_error = -np.inf
else:
multi_phase_error = 0
if activity_kwargs is not None and False:
actvity_error = calculate_activity_error(parameters=parameters, **activity_kwargs)
else:
actvity_error = 0
if thermochemical_kwargs is not None:
single_phase_error = calculate_thermochemical_error(parameters=parameters, **thermochemical_kwargs)
else:
single_phase_error = 0
total_error = multi_phase_error + single_phase_error + actvity_error
logging.log(TRACE, 'Likelihood - {:0.2f}s - Thermochemical: {:0.3f}. ZPF: {:0.3f}. Activity: {:0.3f}. Total: {:0.3f}.'.format(time.time() - starttime, single_phase_error, multi_phase_error, actvity_error, total_error))
lnlike = np.array(total_error, dtype=np.float64)
lnprob = lnprior + lnlike
logging.log(TRACE, 'Proposal - lnprior: {:0.4f}, lnlike: {:0.4f}, lnprob: {:0.4f}'.format(lnprior, lnlike, lnprob))
return lnprob
def predict(params, ctx):
parameters = {
param_name: param
for param_name, param in zip(ctx['symbols_to_fit'], params)
}
zpf_kwargs = ctx.get('zpf_kwargs')
activity_kwargs = ctx.get('activity_kwargs')
thermochemical_kwargs = ctx.get('thermochemical_kwargs')
starttime = time.time()
if zpf_kwargs is not None:
try:
multi_phase_error = calculate_zpf_error(parameters=parameters,
**zpf_kwargs)
except (ValueError, np.linalg.LinAlgError) as e:
raise e
print(e)
multi_phase_error = -np.inf
else:
multi_phase_error = 0
if activity_kwargs is not None:
actvity_error = calculate_activity_error(parameters=parameters,
**activity_kwargs)
else:
actvity_error = 0
if thermochemical_kwargs is not None:
single_phase_error = calculate_thermochemical_error(
parameters=parameters, **thermochemical_kwargs)
else:
single_phase_error = 0
total_error = multi_phase_error + single_phase_error + actvity_error
logging.log(
TRACE,
'Likelihood - {:0.2f}s - Thermochemical: {:0.3f}. ZPF: {:0.3f}. Activity: {:0.3f}. Total: {:0.3f}.'
.format(time.time() - starttime, single_phase_error,
multi_phase_error, actvity_error, total_error))
error = np.array(total_error, dtype=np.float64)
return error
def test_mixing_energies_are_fit(datasets_db):
"""Tests that given mixing energy data, the excess parameter is fit."""
phase_models = {
"components": ["AL", "B"],
"phases": {
"LIQUID": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
},
"FCC_A1": {
"sublattice_model": [["AL", "B"]],
"sublattice_site_ratios": [1]
}
}
}
dataset_excess_mixing = {
"components": ["AL", "B"],
"phases": ["FCC_A1"],
"solver": {
"sublattice_site_ratios": [1],
"sublattice_occupancies": [[[0.5, 0.5]]],
"sublattice_configurations": [[["AL", "B"]]],
"mode": "manual"
},
"conditions": {
"P": 101325,
"T": 298.15
},
"output": "HM_MIX",
"values": [[[-10000]]]
}
datasets_db.insert(dataset_excess_mixing)
dbf = generate_parameters(phase_models, datasets_db, 'SGTE91', 'linear')
assert dbf.elements == {'AL', 'B'}
assert set(dbf.phases.keys()) == {'LIQUID', 'FCC_A1'}
assert len(dbf._parameters.search(where('parameter_type') == 'L')) == 1
assert dbf.symbols['VV0000'] == -40000
# check that read/write is ok
read_dbf = dbf.from_string(dbf.to_string(fmt='tdb'), fmt='tdb')
assert read_dbf.elements == {'AL', 'B'}
assert set(read_dbf.phases.keys()) == {'LIQUID', 'FCC_A1'}
assert len(
read_dbf._parameters.search(where('parameter_type') == 'L')) == 1
from espei.error_functions import calculate_thermochemical_error, get_thermochemical_data
# the error should be exactly 0 because we are only fitting to one point
zero_error_prob = scipy.stats.norm(loc=0, scale=500.0).logpdf(
0.0) # HM weight = 500
# Explicitly pass parameters={} to not try fitting anything
thermochemical_data = get_thermochemical_data(dbf,
sorted(read_dbf.elements),
list(read_dbf.phases.keys()),
datasets_db,
symbols_to_fit=[])
error = calculate_thermochemical_error(dbf, sorted(read_dbf.elements),
thermochemical_data)
assert np.isclose(error, zero_error_prob, atol=1e-6)
