from example_optimization import florisEval
print(
florisEval(
**{
'asv':
1,
'functions':
1,
'cv': [
-2.4120827810e-01, 4.2535457527e-01, -1.1014888289e-01,
-2.2816890223e-01, 2.6956808674e-01, 3.0644245310e-01,
8.2766227014e-02, 1.0617743884e+00, 11.982801151e+00,
8.7694255619e+00
]
}))
#for i, bit in enumerate(results["obj_fn"].asv):
# if bit:
# active_set_vector += 1 << i
# set a dictionary for passing to rosenbrock via Python kwargs
rosen_params = {}
rosen_params['cv'] = continuous_vars
rosen_params['asv'] = [active_set_vector]
rosen_params['functions'] = 1
# execute the rosenbrock analysis as a separate Python module
#from rosenbrock import rosenbrock_list
from example_optimization import florisEval
if np.isinf(params['direction']): rosen_results = {'fns': [0, 0, 0]}
else: rosen_results = florisEval(**rosen_params)
# ----------------------------
# Return the results to Dakota
# ----------------------------
# Insert functions from rosen into results
# Results iterator provides an index, response name, and response
for i, n, r in results:
if r.asv.function:
r.function = rosen_results['fns'][i]
if r.asv.gradient:
r.gradient = rosen_results['fnGrads'][i]
if r.asv.hessian:
r.hessian = rosen_results['fnHessians'][i]
import sys
from example_optimization import florisEval
import numpy as np
yaw = np.genfromtxt('./%s.dat' % sys.argv[-1])
f = str((florisEval(**{
'functions': -33,
'asv': 1e3,
'cv': np.array(yaw)
}))['fns'][0])
outf = open('%s.det' % sys.argv[-1], 'w')
outf.write(f)
outf.close()