Reputation: 176
This is an add-on to a previous question about iterating the values in and openMDAO problem. I want to be able to calculate values with top.run(). Then inspect the changes. However the previous advice does not seem to work.
I have found that prob.root.unknowns does not really change when the problem is modified or ran.
class TestObj(Component):
def __init__(self):
Component.__init__(self)
self.add_param('x',5.0)
self.add_output('y',5.0)
def solve_nonlinear(self,params,unknowns,resids):
unknowns['y']=params['x']
top = Problem()
root = Group()
top.root=root
root.add('split',IndepVarComp('x',2.0),promotes=['*'])
root.add('test',TestObj(),promotes=['*'])
top.setup()
top.root.list_connections()
print 'starting'
print top['x']
print top['y']
for k,v in top.root.unknowns.iteritems():
print k
print v['val']
top.run()
print 'after first solve'
print top['x']
print top['y']
for k,v in top.root.unknowns.iteritems():
print k
print v['val']
top['x']=33.0
print 'after second solve'
print top['x']
print top['y']
for k,v in top.root.unknowns.iteritems():
print k
print v['val']
top.run()
print 'after third solve'
print top['x']
print top['y']
for k,v in top.root.unknowns.iteritems():
print k
print v['val']
Upvotes: 0
Views: 28
Reputation: 5710
The iterator over a vec_wrapper returns an iterator over the keys, and the meta-data dictionary. The 'val' attribute you're accessing in the meta-data is actually the initial value.
Try this instead, and you'll see that it is changing:
from openmdao.api import Component, Problem, Group, IndepVarComp
class TestObj(Component):
def __init__(self):
Component.__init__(self)
self.add_param('x',5.0)
self.add_output('y',5.0)
def solve_nonlinear(self,params,unknowns,resids):
unknowns['y']=params['x']
top = Problem()
root = Group()
top.root=root
root.add('split',IndepVarComp('x',2.0),promotes=['*'])
root.add('test',TestObj(),promotes=['*'])
top.setup()
top.root.list_connections()
print 'starting'
print top['x']
print top['y']
for k,v in root.unknowns.iteritems():
print k
print top.root.unknowns[k]
top.run()
print 'after first solve'
print top['x']
print top['y']
for k,v in root.unknowns.iteritems():
print k
print top.root.unknowns[k]
top['x']=33.0
print 'after second solve'
print top['x']
print top['y']
for k,v in top.root.unknowns.iteritems():
print k
print top.root.unknowns[k]
top.run()
print 'after third solve'
print top['x']
print top['y']
for k,v in top.root.unknowns.iteritems():
print k
print top.root.unknowns[k]
Upvotes: 1