Address a numpy deprecation warning about assigning a vector to a single vector index.

[Kenneth-T-Moore]

Summary

Address a numpy deprecation warning about assigning a vector to a single vector index.

Related Issues

• Resolves #

Backwards incompatibilities

None

New Dependencies

None

[jkirk5]

Do we know why those variables are vectors when they shouldn't be? I'm fine with this fix but we might run into this issue in other places if we don't address the root cause

[Kenneth-T-Moore]

Do we know why those variables are vectors when they shouldn't be? I'm fine with this fix but we might run into this issue in other places if we don't address the root cause

It seems to mainly be openmdao behavior right now that everything you get from inputs is a vector, no matter how you declare it.

    import numpy as np
    
    import openmdao.api as om
    
    class MyComp(om.ExplicitComponent):
    
        def setup(self):
    
            self.add_input('x1', val = 1.0)
            self.add_input('x2', val = np.array((1, )))
            self.add_input('x3', shape=1)
            self.add_input('x4', shape=(1, ))
            self.add_input('x5', shape=0)
            self.add_output('zzz')
    
        def compute(self, inputs, outputs, discrete_inputs=None, discrete_outputs=None):
            
            print('x1', inputs['x1'])
            print('x2', inputs['x2'])
            print('x3', inputs['x3'])
            print('x4', inputs['x4'])
            print('x5', inputs['x5'])
    
    prob = om.Problem()
    model = prob.model
    model.add_subsystem('comp', MyComp())
    prob.setup()
    
    prob.run_model()
    print('done')

/Users/ktmoore1/Work/OpenMDAO/OpenMDAO/openmdao/visualization/n2_viewer/n2_viewer.py:114: OpenMDAOWarning:All-NaN slice encountered
x1 [1.]
x2 [1.]
x3 [1.]
x4 [1.]
x5 []
done