def test_add_scan_variable(self):
attrs = {'type': 'scan'}
lower = [0.1, 0.2, 0.3, 0.4]
upper = [1.0, 2.0, 3.0, 4.0]
partitions = [2, 3, 4, 5]
lower = xr.DataArray(data=lower, dims='T')
upper = xr.DataArray(data=upper, dims='T')
partitions = xr.DataArray(data=partitions, dims='T')
test_var = xr.Dataset(
{
'lower_bounds': lower,
'upper_bounds': upper,
'partitions': partitions
},
attrs=attrs)
my_dakota = DakotaClass()
my_dakota.add_variable('test_var', test_var)
self.assertTrue(
np.array_equal(my_dakota.dakota.get_attribute('lower_bounds'),
lower))
self.assertTrue(
np.array_equal(my_dakota.dakota.get_attribute('upper_bounds'),
upper))
def test_add_variable_missing_data(self):
attrs = {'type': 'normal'}
means = [1.0, 2.0, 3.0, 4.0]
test_var = xr.Dataset({'means': means}, attrs=attrs)
my_dakota = DakotaClass()
with self.assertRaises(DatasetError):
my_dakota.add_variable('test_var', test_var)
def test_add_variable_not_dataset(self):
means = [1.0, 2.0, 3.0, 4.0]
sds = [0.1, 0.2, 0.3, 0.4]
test_var = {'means': means, 'std_deviations': sds}
my_dakota = DakotaClass()
with self.assertRaises(DatasetError):
my_dakota.add_variable('test_var', test_var)
def test_add_run_settings(self):
attrs = {'sample_type': 'sampling', 'seed': 54}
new_settings = xr.Dataset(attrs=attrs)
my_dakota = DakotaClass()
my_dakota.update_settings(new_settings)
self.assertEqual(
my_dakota.dakota.get_attribute('sample_type').strip(), 'sampling')
self.assertEqual(my_dakota.dakota.get_attribute('seed'), 54)
def test_add_variable_with_no_type(self):
means = [1.0, 2.0, 3.0, 4.0]
sds = [0.1, 0.2, 0.3, 0.4]
means = xr.DataArray(data=means, dims='T')
sds = xr.DataArray(data=sds, dims='T')
test_var = xr.Dataset({'means': means, 'std_deviations': sds})
my_dakota = DakotaClass()
with self.assertRaises(DatasetError):
my_dakota.add_variable('test_var', test_var)
def test_add_variable_with_nans(self):
attrs = {'type': 'normal'}
means = [1.0, 2.0, np.nan, 4.0]
sds = [0.1, 0.2, 0.3, 0.4]
test_var = xr.Dataset({
'means': means,
'std_deviations': sds
},
attrs=attrs)
my_dakota = DakotaClass()
with self.assertRaises(DatasetError):
my_dakota.add_variable('test_var', test_var)
def test_create_dakota_template(self):
my_dakota = DakotaClass()
self.assertEqual(
my_dakota.dakota.get_attribute('evaluation_concurrency'), 1)
self.assertEqual(my_dakota.dakota.get_attribute('response_functions'),
1)
def make_dakota(args):
# Create template DAKOTA input file
dakota = DakotaClass()
# Read user provided file
user_file = DakotaFile(file_type=args.type)
user_file.read(args.input)
# Set Dakota entries from arguments
dakota.dakota.set_attribute('evaluation_concurrency', args.concurrency)
dakota.dakota.set_attribute('analysis_drivers', "'" + args.driver + "'")
# Update DAKOTA input file with uncertain variables
user_file.write_dakota_input(dakota)
# Write input file
dakota.write_input_file(args.output)
def test_add_lognormal_variable(self):
attrs = {'type': 'lognormal'}
means = [1.0, 2.0, 3.0, 4.0]
sds = [0.1, 0.2, 0.3, 0.4]
means = xr.DataArray(data=means, dims='T')
sds = xr.DataArray(data=sds, dims='T')
test_var = xr.Dataset({
'means': means,
'std_deviations': sds
},
attrs=attrs)
my_dakota = DakotaClass()
my_dakota.add_variable('test_var', test_var)
self.assertTrue(
np.array_equal(my_dakota.dakota.get_attribute('means'), means))
self.assertTrue(
np.array_equal(my_dakota.dakota.get_attribute('std_deviations'),
sds))
def test_add_correlated_scan_variable_with_inconsistent_partitions(self):
attrs = {'type': 'scan_correlated'}
lower = [0.1, 0.2, 0.3, 0.4]
upper = [1.0, 2.0, 3.0, 4.0]
partitions = [4, 5, 4, 4]
lower = xr.DataArray(data=lower, dims='T')
upper = xr.DataArray(data=upper, dims='T')
partitions = xr.DataArray(data=partitions, dims='T')
test_var = xr.Dataset(
{
'lower_bounds': lower,
'upper_bounds': upper,
'partitions': partitions
},
attrs=attrs)
my_dakota = DakotaClass()
with self.assertRaises(DatasetError):
my_dakota.add_variable('test_var', test_var)
def test_write_dakote_file(self):
my_dakota = DakotaClass()
my_dakota.write_input_file('test_dakota.dat')
self.assertTrue(os.path.isfile('test_dakota.dat'))
os.remove('test_dakota.dat')