def test_generate_numpydocstring(self):
group = Group()
group.add('x_param', IndepVarComp('x', 1.0), promotes=['*'])
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem()
prob.root = group
prob.root.ln_solver = ScipyGMRES()
test_string = prob.root.ln_solver.generate_docstring()
original_string = \
""" \"\"\"
Options
-------
options['atol'] : float(1e-12)
Absolute convergence tolerance.
options['err_on_maxiter'] : bool(False)
If True, raise an AnalysisError if not converged at maxiter.
options['iprint'] : int(0)
Set to 0 to disable printing, set to 1 to print the residual to stdout each iteration, set to 2 to print subiteration residuals as well.
options['maxiter'] : int(1000)
Maximum number of iterations.
options['mode'] : str('auto')
Derivative calculation mode, set to 'fwd' for forward mode, 'rev' for reverse mode, or 'auto' to let OpenMDAO determine the best mode.
options['restart'] : int(20)
Number of iterations between restarts. Larger values increase iteration cost, but may be necessary for convergence
\"\"\"
"""
for sorig, stest in zip(original_string.split('\n'),
test_string.split('\n')):
self.assertEqual(sorig, stest)
def test_generate_numpydocstring(self):
group = Group()
group.add('x_param', IndepVarComp('x', 1.0), promotes=['*'])
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem()
prob.root = group
prob.root.ln_solver = ScipyGMRES()
test_string = prob.root.ln_solver.generate_docstring()
original_string = ' """\n\n Options\n -------\n options[\'atol\'] : float(1e-12)\n Absolute convergence tolerance.\n options[\'iprint\'] : int(0)\n Set to 0 to disable printing, set to 1 to print the residual to stdout each iteration, set to 2 to print subiteration residuals as well.\n options[\'maxiter\'] : int(1000)\n Maximum number of iterations.\n options[\'mode\'] : str(\'auto\')\n Derivative calculation mode, set to \'fwd\' for forward mode, \'rev\' for reverse mode, or \'auto\' to let OpenMDAO determine the best mode.\n options[\'precondition\'] : bool(False)\n Set to True to turn on preconditioning.\n options[\'restart\'] : int(20)\n Number of iterations between restarts. Larger values increase iteration cost, but may be necessary for convergence\n\n """\n'
self.assertEqual(original_string, test_string)
def test_simple_matvec(self):
group = Group()
group.add('x_param', IndepVarComp('x', 1.0), promotes=['*'])
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem()
prob.root = group
prob.root.ln_solver = LinearGaussSeidel()
prob.setup(check=False)
prob.run()
J = prob.calc_gradient(['x'], ['y'], mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='rev', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
def test_simple_in_group_matvec(self):
group = Group()
sub = group.add('sub', Group(), promotes=['x', 'y'])
group.add('x_param', ParamComp('x', 1.0), promotes=['*'])
sub.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem()
prob.root = group
prob.root.ln_solver = ExplicitSolver()
prob.setup(check=False)
prob.run()
J = prob.calc_gradient(['x'], ['y'], mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='rev', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
def test_simple_choose_different_alg(self):
group = Group()
group.add('x_param', IndepVarComp('x', 1.0), promotes=['*'])
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem(impl=impl)
prob.root = group
prob.root.ln_solver = PetscKSP()
prob.root.ln_solver.options['ksp_type'] = 'gmres'
prob.setup(check=False)
prob.run()
J = prob.calc_gradient(['x'], ['y'], mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='rev', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
def test_full_model_fd_simple_comp_promoted(self):
prob = Problem()
prob.root = Group()
sub = prob.root.add('sub', Group(), promotes=['*'])
sub.add('comp', SimpleCompDerivMatVec(), promotes=['*'])
prob.root.add('p1', IndepVarComp('x', 1.0), promotes=['*'])
prob.root.fd_options['force_fd'] = True
prob.setup(check=False)
prob.run()
indep_list = ['x']
unknown_list = ['y']
J = prob.calc_gradient(indep_list, unknown_list, mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
def test_simple_matvec(self):
group = Group()
group.add('x_param', IndepVarComp('x', 1.0), promotes=['*'])
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem()
prob.root = group
prob.root.ln_solver = DirectSolver()
prob.root.ln_solver.options['jacobian_method'] = 'assemble'
prob.setup(check=False)
prob.run()
with self.assertRaises(RuntimeError) as cm:
J = prob.calc_gradient(['x'], ['y'], mode='fwd', return_format='dict')
expected_msg = "The 'assemble' jacobian_method is not supported when " + \
"'apply_linear' is used on a component (mycomp)."
self.assertEqual(str(cm.exception), expected_msg)
def test_full_model_fd_simple_comp_promoted(self):
prob = Problem()
prob.root = Group()
sub = prob.root.add('sub', Group(), promotes=['*'])
sub.add('comp', SimpleCompDerivMatVec(), promotes=['*'])
prob.root.add('p1', IndepVarComp('x', 1.0), promotes=['*'])
prob.setup(check=False)
prob.run()
data = prob.check_total_derivatives(out_stream=None)
for key, val in iteritems(data):
assert_rel_error(self, val['abs error'][0], 0.0, 1e-5)
assert_rel_error(self, val['abs error'][1], 0.0, 1e-5)
assert_rel_error(self, val['abs error'][2], 0.0, 1e-5)
assert_rel_error(self, val['rel error'][0], 0.0, 1e-5)
assert_rel_error(self, val['rel error'][1], 0.0, 1e-5)
assert_rel_error(self, val['rel error'][2], 0.0, 1e-5)
def test_simple_matvec_subbed(self):
group = Group()
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem(impl=impl)
prob.root = Group()
prob.root.add('x_param', ParamComp('x', 1.0), promotes=['*'])
prob.root.add('sub', group, promotes=['*'])
prob.root.ln_solver = PetscKSP()
prob.setup(check=False)
prob.run()
J = prob.calc_gradient(['x'], ['y'], mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='rev', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='fd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
def test_full_model_fd_simple_comp(self):
prob = Problem()
prob.root = Group()
prob.root.add('comp', SimpleCompDerivMatVec())
prob.root.add('p1', ParamComp('x', 1.0))
prob.root.connect('p1.x', 'comp.x')
prob.root.fd_options['force_fd'] = True
prob.setup(check=False)
prob.run()
param_list = ['comp.x']
unknown_list = ['comp.y']
J = prob.calc_gradient(param_list,
unknown_list,
mode='fwd',
return_format='dict')
assert_rel_error(self, J['comp.y']['comp.x'][0][0], 2.0, 1e-6)
def test_full_model_fd_simple_comp_promoted(self):
prob = Problem()
prob.root = Group()
sub = prob.root.add('sub', Group(), promotes=['*'])
sub.add('comp', SimpleCompDerivMatVec(), promotes=['*'])
prob.root.add('p1', IndepVarComp('x', 1.0), promotes=['*'])
prob.root.deriv_options['type'] = 'fd'
prob.setup(check=False)
prob.run()
indep_list = ['x']
unknown_list = ['y']
J = prob.calc_gradient(indep_list, unknown_list, mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
# We should not allocate deriv vectors for full model FD
self.assertEqual(len(prob.root.dumat[None].vec), 0)
self.assertEqual(len(prob.root.drmat[None].vec), 0)
self.assertEqual(len(prob.root.dpmat[None].vec), 0)
def test_simple_matvec_subbed_like_multipoint(self):
group = Group()
group.add('mycomp', SimpleCompDerivMatVec(), promotes=['x', 'y'])
prob = Problem()
prob.root = Group()
prob.root.add('sub', group, promotes=['*'])
prob.root.sub.add('x_param', IndepVarComp('x', 1.0), promotes=['*'])
prob.root.ln_solver = ScipyGMRES()
prob.setup(check=False)
prob.run()
J = prob.calc_gradient(['x'], ['y'], mode='fwd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='rev', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='fd', return_format='dict')
assert_rel_error(self, J['y']['x'][0][0], 2.0, 1e-6)
J = prob.calc_gradient(['x'], ['y'], mode='fd', return_format='array')
assert_rel_error(self, J[0][0], 2.0, 1e-6)
def test_full_model_fd_simple_comp(self):
prob = Problem()
prob.root = Group()
prob.root.add('comp', SimpleCompDerivMatVec())
prob.root.add('p1', IndepVarComp('x', 1.0))
prob.root.connect('p1.x', 'comp.x')
prob.root.fd_options['force_fd'] = True
prob.setup(check=False)
prob.run()
indep_list = ['comp.x']
unknown_list = ['comp.y']
J = prob.calc_gradient(indep_list, unknown_list, mode='fwd', return_format='dict')
assert_rel_error(self, J['comp.y']['comp.x'][0][0], 2.0, 1e-6)
# We should not allocate deriv vectors for full model FD
self.assertEqual(len(prob.root.dumat[None].vec), 0)
self.assertEqual(len(prob.root.drmat[None].vec), 0)
self.assertEqual(len(prob.root.dpmat[None].vec), 0)
