def _apply_to(self, instance, **kwds):
if not instance.ctype in (Block, Disjunct):
raise GDP_Error(
"Transformation called on %s of type %s. 'instance'"
" must be a ConcreteModel, Block, or Disjunct (in "
"the case of nested disjunctions)." %
(instance.name, instance.ctype))
try:
self._config = self.CONFIG(kwds.pop('options', {}))
self._config.set_value(kwds)
self._transformation_blocks = {}
if not self._config.assume_fixed_vars_permanent:
fixed_vars = ComponentMap()
for v in get_vars_from_components(instance,
Constraint,
include_fixed=True,
active=True,
descend_into=(Block,
Disjunct)):
if v.fixed:
fixed_vars[v] = value(v)
v.fixed = False
self._apply_to_impl(instance)
finally:
# restore fixed variables
if not self._config.assume_fixed_vars_permanent:
for v, val in fixed_vars.items():
v.fix(val)
del self._config
del self._transformation_blocks
def test_eval_numpy(self):
m = ConcreteModel()
m.p = Param([1,2], mutable=True)
m.x = Var()
data = np.array([[0,-1,2],
[.1,.2,.3],
[4,5,6]])
cMap = ComponentMap()
cMap[m.p[1]] = data[0]
cMap[m.p[2]] = data[1]
cMap[m.x] = data[2]
npe = NumpyEvaluator(cMap)
result = npe.walk_expression(sin(m.x))
self.assertEqual(result[0], sin(4))
self.assertEqual(result[1], sin(5))
self.assertEqual(result[2], sin(6))
result = npe.walk_expression(abs(m.x * m.p[1] - m.p[2]))
self.assertEqual(result[0], .1)
self.assertEqual(result[1], -((-1*5)-.2))
self.assertEqual(result[2], (2*6-.3))
result = npe.walk_expression(atan(m.x))
self.assertEqual(result[0], atan(4))
self.assertEqual(result[1], atan(5))
self.assertEqual(result[2], atan(6))
result = npe.walk_expression(atanh(m.p[2]))
self.assertEqual(result[0], atanh(.1))
self.assertEqual(result[1], atanh(.2))
self.assertEqual(result[2], atanh(.3))
def test_eval_numpy(self):
m = ConcreteModel()
m.p = Param([1,2], mutable=True)
m.x = Var()
data = np.array([[0,-1,2],
[.1,.2,.3],
[4,5,6]])
cMap = ComponentMap()
cMap[m.p[1]] = data[0]
cMap[m.p[2]] = data[1]
cMap[m.x] = data[2]
npe = NumpyEvaluator(cMap)
result = npe.walk_expression(sin(m.x))
assert pytest.approx(result[0], rel=1e-12) == sin(4)
assert pytest.approx(result[1], rel=1e-12) == sin(5)
assert pytest.approx(result[2], rel=1e-12) == sin(6)
result = npe.walk_expression(abs(m.x * m.p[1] - m.p[2]))
assert pytest.approx(result[0], rel=1e-12) == .1
assert pytest.approx(result[1], rel=1e-12) == -((-1*5)-.2)
assert pytest.approx(result[2], rel=1e-12) == (2*6-.3)
result = npe.walk_expression(atan(m.x))
assert pytest.approx(result[0], rel=1e-12) == atan(4)
assert pytest.approx(result[1], rel=1e-12) == atan(5)
assert pytest.approx(result[2], rel=1e-12) == atan(6)
result = npe.walk_expression(atanh(m.p[2]))
assert pytest.approx(result[0], rel=1e-12) == atanh(.1)
assert pytest.approx(result[1], rel=1e-12) == atanh(.2)
assert pytest.approx(result[2], rel=1e-12) == atanh(.3)
def test_improved_bounds(self):
m = ConcreteModel()
m.x = Var(bounds=(0, 100), initialize=5)
improved_bounds = ComponentMap()
improved_bounds[m.x] = (10, 20)
mc_expr = mc(m.x, improved_var_bounds=improved_bounds)
self.assertEqual(mc_expr.lower(), 10)
self.assertEqual(mc_expr.upper(), 20)
def disjunctive_bounds(scope):
"""Return all of the variable bounds defined at a disjunctive scope."""
possible_disjunct = scope
while possible_disjunct is not None:
try:
return possible_disjunct._disj_var_bounds
except AttributeError:
# possible disjunct does not have attribute '_disj_var_bounds'.
# Try again with the scope's parent block.
possible_disjunct = possible_disjunct.parent_block()
# Unable to find '_disj_var_bounds' attribute within search scope.
return ComponentMap()
def test_eval_constant(self):
m = ConcreteModel()
m.p = Param([1,2], mutable=True)
m.x = Var(initialize=0.25)
cMap = ComponentMap()
cMap[m.p[1]] = 2
cMap[m.p[2]] = 4
npe = NumpyEvaluator(cMap)
expr = m.p[1] + m.p[2] + m.x + .5
self.assertEqual(npe.walk_expression(expr), 6.75)
m.p[1] = 2
m.p[2] = 4
self.assertEqual(value(expr), 6.75)
def test_eval_constant(self):
m = ConcreteModel()
m.p = Param([1,2], mutable=True)
m.x = Var(initialize=0.25)
cMap = ComponentMap()
cMap[m.p[1]] = 2
cMap[m.p[2]] = 4
npe = NumpyEvaluator(cMap)
expr = m.p[1] + m.p[2] + m.x + .5
assert npe.walk_expression(expr) == pytest.approx(6.75, rel=1e-12)
m.p[1] = 2
m.p[2] = 4
assert value(expr) == pytest.approx(6.75, rel=1e-12)
def __init__(self, sVar, **kwds):
if not isinstance(sVar, Var):
raise DAE_Error(
"%s is not a variable. Can only take the derivative of a Var"
"component." % sVar)
if "wrt" in kwds and "withrespectto" in kwds:
raise TypeError(
"Cannot specify both 'wrt' and 'withrespectto keywords "
"in a DerivativeVar")
wrt = kwds.pop('wrt', None)
wrt = kwds.pop('withrespectto', wrt)
try:
num_contset = len(sVar._contset)
except AttributeError:
# This dictionary keeps track of where the ContinuousSet appears
# in the index. This implementation assumes that every element
# in an indexing set has the same dimension.
sVar._contset = ComponentMap()
sVar._derivative = {}
if sVar.dim() == 0:
num_contset = 0
else:
sidx_sets = list(sVar.index_set().subsets())
loc = 0
for i, s in enumerate(sidx_sets):
if s.type() is ContinuousSet:
sVar._contset[s] = loc
_dim = s.dimen
if _dim is None:
raise DAE_Error(
"The variable %s is indexed by a Set (%s) with a "
"non-fixed dimension. A DerivativeVar may only be "
"indexed by Sets with constant dimension" %
(sVar, s.name))
elif _dim is UnknownSetDimen:
raise DAE_Error(
"The variable %s is indexed by a Set (%s) with an "
"unknown dimension. A DerivativeVar may only be "
"indexed by Sets with known constant dimension" %
(sVar, s.name))
loc += s.dimen
num_contset = len(sVar._contset)
if num_contset == 0:
raise DAE_Error(
"The variable %s is not indexed by any ContinuousSets. A "
"derivative may only be taken with respect to a continuous "
"domain" % sVar)
if wrt is None:
# Check to be sure Var is indexed by single ContinuousSet and take
# first deriv wrt that set
if num_contset != 1:
raise DAE_Error(
"The variable %s is indexed by multiple ContinuousSets. "
"The desired ContinuousSet must be specified using the "
"keyword argument 'wrt'" % sVar)
wrt = [
next(iterkeys(sVar._contset)),
]
elif type(wrt) is ContinuousSet:
if wrt not in sVar._contset:
raise DAE_Error(
"Invalid derivative: The variable %s is not indexed by "
"the ContinuousSet %s" % (sVar, wrt))
wrt = [
wrt,
]
elif type(wrt) is tuple or type(wrt) is list:
for i in wrt:
if type(i) is not ContinuousSet:
raise DAE_Error(
"Cannot take the derivative with respect to %s. "
"Expected a ContinuousSet or a tuple of "
"ContinuousSets" % i)
if i not in sVar._contset:
raise DAE_Error(
"Invalid derivative: The variable %s is not indexed "
"by the ContinuousSet %s" % (sVar, i))
wrt = list(wrt)
else:
raise DAE_Error(
"Cannot take the derivative with respect to %s. "
"Expected a ContinuousSet or a tuple of ContinuousSets" % i)
wrtkey = [str(i) for i in wrt]
wrtkey.sort()
wrtkey = tuple(wrtkey)
if wrtkey in sVar._derivative:
raise DAE_Error(
"Cannot create a new derivative variable for variable "
"%s: derivative already defined as %s" %
(sVar.name, sVar._derivative[wrtkey]().name))
sVar._derivative[wrtkey] = weakref.ref(self)
self._sVar = sVar
self._wrt = wrt
kwds.setdefault('ctype', DerivativeVar)
Var.__init__(self, sVar.index_set(), **kwds)
def generate_structured_model(self):
"""
Using the community map and the original model used to create this community map, we will create
structured_model, which will be based on the original model but will place variables, constraints, and
objectives into or outside of various blocks (communities) based on the community map.
Returns
-------
structured_model: Block
a Pyomo model that reflects the nature of the community map
"""
# Initialize a new model (structured_model) which will contain variables and constraints in blocks based on
# their respective communities within the CommunityMap
structured_model = ConcreteModel()
# Create N blocks (where N is the number of communities found within the model)
structured_model.b = Block([0, len(self.community_map) - 1,
1]) # values given for (start, stop, step)
# Initialize a ComponentMap that will map a variable from the model (for example, old_model.x1) used to
# create the CommunityMap to a list of variables in various blocks that were created based on this
# variable (for example, [structured_model.b[0].x1, structured_model.b[3].x1])
blocked_variable_map = ComponentMap()
# Example key-value pair -> {original_model.x1 : [structured_model.b[0].x1, structured_model.b[3].x1]}
# TODO - Consider changing structure of the next two for loops to be more efficient (maybe loop through
# constraints and add variables as you go) (but note that disconnected variables would be
# missed with this strategy)
# First loop through community_map to add all the variables to structured_model before we add constraints
# that use those variables
for community_key, community in self.community_map.items():
_, variables_in_community = community
# Loop through all of the variables (from the original model) in the given community
for stored_variable in variables_in_community:
# Construct a new_variable whose attributes are determined by querying the variable from the
# original model
new_variable = Var(domain=stored_variable.domain,
bounds=stored_variable.bounds)
# Add this new_variable to its block/community and name it using the string of the variable from the
# original model
structured_model.b[community_key].add_component(
str(stored_variable), new_variable)
# Since there could be multiple variables 'x1' (such as
# structured_model.b[0].x1, structured_model.b[3].x1, etc), we need to create equality constraints
# for all of the variables 'x1' within structured_model (this is the purpose of blocked_variable_map)
# Here we update blocked_variable_map to keep track of what equality constraints need to be made
variable_in_new_model = structured_model.find_component(
new_variable)
blocked_variable_map[
stored_variable] = blocked_variable_map.get(
stored_variable, []) + [variable_in_new_model]
# Now that we have all of our variables within the model, we will initialize a dictionary that used to
# replace variables within constraints to other variables (in our case, this will convert variables from the
# original model into variables from the new model (structured_model))
replace_variables_in_expression_map = dict()
# Loop through community_map again, this time to add constraints (with replaced variables)
for community_key, community in self.community_map.items():
constraints_in_community, _ = community
# Loop through all of the constraints (from the original model) in the given community
for stored_constraint in constraints_in_community:
# Now, loop through all of the variables within the given constraint expression
for variable_in_stored_constraint in identify_variables(
stored_constraint.expr):
# Loop through each of the "blocked" variables that a variable is mapped to and update
# replace_variables_in_expression_map if a variable has a "blocked" form in the given community
# What this means is that if we are looping through constraints in community 0, then it would be
# best to change a variable x1 into b[0].x1 as opposed to b[2].x1 or b[5].x1 (assuming all of these
# blocked versions of the variable x1 exist (which depends on the community map))
variable_in_current_block = False
for blocked_variable in blocked_variable_map[
variable_in_stored_constraint]:
if 'b[%d]' % community_key in str(blocked_variable):
# Update replace_variables_in_expression_map accordingly
replace_variables_in_expression_map[
id(variable_in_stored_constraint
)] = blocked_variable
variable_in_current_block = True
if not variable_in_current_block:
# Create a version of the given variable outside of blocks then add it to
# replace_variables_in_expression_map
new_variable = Var(
domain=variable_in_stored_constraint.domain,
bounds=variable_in_stored_constraint.bounds)
# Add the new variable just as we did above (but now it is not in any blocks)
structured_model.add_component(
str(variable_in_stored_constraint), new_variable)
# Update blocked_variable_map to keep track of what equality constraints need to be made
variable_in_new_model = structured_model.find_component(
new_variable)
blocked_variable_map[
variable_in_stored_constraint] = blocked_variable_map.get(
variable_in_stored_constraint,
[]) + [variable_in_new_model]
# Update replace_variables_in_expression_map accordingly
replace_variables_in_expression_map[
id(variable_in_stored_constraint
)] = variable_in_new_model
# TODO - Is there a better way to check whether something is actually an objective? (as done below)
# Check to see whether 'stored_constraint' is actually an objective (since constraints and objectives
# grouped together)
if self.with_objective and isinstance(
stored_constraint, (_GeneralObjectiveData, Objective)):
# If the constraint is actually an objective, we add it to the block as an objective
new_objective = Objective(expr=replace_expressions(
stored_constraint.expr,
replace_variables_in_expression_map))
structured_model.b[community_key].add_component(
str(stored_constraint), new_objective)
else:
# Construct a constraint based on the expression within stored_constraint and the dict we have
# created for the purpose of replacing the variables within the constraint expression
new_constraint = Constraint(expr=replace_expressions(
stored_constraint.expr,
replace_variables_in_expression_map))
# Add this new constraint to the corresponding community/block with its name as the string of the
# constraint from the original model
structured_model.b[community_key].add_component(
str(stored_constraint), new_constraint)
# If with_objective was set to False, that means we might have missed an objective function within the
# original model
if not self.with_objective:
# Construct a new dictionary for replacing the variables (replace_variables_in_objective_map) which will
# be specific to the variables in the objective function, since there is the possibility that the
# objective contains variables we have not yet seen (and thus not yet added to our new model)
for objective_function in self.model.component_data_objects(
ctype=Objective,
active=self.use_only_active_components,
descend_into=True):
for variable_in_objective in identify_variables(
objective_function):
# Add all of the variables in the objective function (not within any blocks)
# Check to make sure a form of the variable has not already been made outside of the blocks
if structured_model.find_component(
str(variable_in_objective)) is None:
new_variable = Var(domain=variable_in_objective.domain,
bounds=variable_in_objective.bounds)
structured_model.add_component(
str(variable_in_objective), new_variable)
# Again we update blocked_variable_map to keep track of what
# equality constraints need to be made
variable_in_new_model = structured_model.find_component(
new_variable)
blocked_variable_map[
variable_in_objective] = blocked_variable_map.get(
variable_in_objective,
[]) + [variable_in_new_model]
# Update the dictionary that we will use to replace the variables
replace_variables_in_expression_map[id(
variable_in_objective)] = variable_in_new_model
else:
for version_of_variable in blocked_variable_map[
variable_in_objective]:
if 'b[' not in str(version_of_variable):
replace_variables_in_expression_map[
id(variable_in_objective
)] = version_of_variable
# Now we will construct a new objective function based on the one from the original model and then
# add it to the new model just as we have done before
new_objective = Objective(expr=replace_expressions(
objective_function.expr,
replace_variables_in_expression_map))
structured_model.add_component(str(objective_function),
new_objective)
# Now, we need to create equality constraints for all of the different "versions" of a variable (such
# as x1, b[0].x1, b[2].x2, etc.)
# Create a constraint list for the equality constraints
structured_model.equality_constraint_list = ConstraintList(
doc="Equality Constraints for the different "
"forms of a given variable")
# Loop through blocked_variable_map and create constraints accordingly
for variable, duplicate_variables in blocked_variable_map.items():
# variable -> variable from the original model
# duplicate_variables -> list of variables in the new model
# Create a list of all the possible equality constraints that need to be made
equalities_to_make = combinations(duplicate_variables, 2)
# Loop through the list of two-variable tuples and create an equality constraint for those two variables
for variable_1, variable_2 in equalities_to_make:
structured_model.equality_constraint_list.add(
expr=variable_1 == variable_2)
# Return 'structured_model', which is essentially identical to the original model but now has all of the
# variables, constraints, and objectives placed into blocks based on the nature of the CommunityMap
return structured_model
def visualize_model_graph(self,
type_of_graph='constraint',
filename=None,
pos=None):
"""
This function draws a graph of the communities for a Pyomo model.
The type_of_graph parameter is used to create either a variable-node graph, constraint-node graph, or
bipartite graph of the Pyomo model. Then, the nodes are colored based on the communities they are in - which
is based on the community map (self.community_map). A filename can be provided to save the figure, otherwise
the figure is illustrated with matplotlib.
Parameters
----------
type_of_graph: str, optional
a string that specifies the types of nodes drawn on the model graph, the default is 'constraint'.
'constraint' draws a graph with constraint nodes,
'variable' draws a graph with variable nodes,
'bipartite' draws a bipartite graph (with both constraint and variable nodes)
filename: str, optional
a string that specifies a path for the model graph illustration to be saved
pos: dict, optional
a dictionary that maps node keys to their positions on the illustration
Returns
-------
fig: matplotlib figure
the figure for the model graph drawing
pos: dict
a dictionary that maps node keys to their positions on the illustration - can be used to create consistent
layouts for graphs of a given model
"""
# Check that all arguments are of the correct type
assert type_of_graph in ('bipartite', 'constraint', 'variable'), \
"Invalid graph type specified: 'type_of_graph=%s' - Valid values: " \
"'bipartite', 'constraint', 'variable'" % type_of_graph
assert isinstance(filename, (type(None), str)), "Invalid value for filename: 'filename=%s' - filename " \
"must be a string" % filename
# No assert statement for pos; the NetworkX function can handle issues with the pos argument
# There is a possibility that the desired networkX graph of the model is already stored in the
# CommunityMap object (because the networkX graph is required to create the CommunityMap object)
if type_of_graph != self.type_of_community_map:
# Use the generate_model_graph function to create a NetworkX graph of the given model (along with
# number_component_map and constraint_variable_map, which will be used to help with drawing the graph)
model_graph, number_component_map, constraint_variable_map = generate_model_graph(
self.model,
type_of_graph=type_of_graph,
with_objective=self.with_objective,
weighted_graph=self.weighted_graph,
use_only_active_components=self.use_only_active_components)
else:
# This is the case where, as mentioned above, we can use the networkX graph that was made to create
# the CommunityMap object
model_graph, number_component_map, constraint_variable_map = self.graph, self.graph_node_mapping, \
self.constraint_variable_map
# This line creates the "reverse" of the number_component_map above, since mapping the Pyomo
# components to their nodes in the networkX graph is more convenient in this function
component_number_map = ComponentMap(
(comp, number) for number, comp in number_component_map.items())
# Create a deep copy of the community_map attribute to avoid destructively modifying it
numbered_community_map = copy.deepcopy(self.community_map)
# Now we will use the component_number_map to change the Pyomo modeling components in community_map into the
# numbers that correspond to their nodes/edges in the NetworkX graph, model_graph
for key in self.community_map:
numbered_community_map[key] = ([
component_number_map[component]
for component in self.community_map[key][0]
], [
component_number_map[component]
for component in self.community_map[key][1]
])
# Based on type_of_graph, which specifies what Pyomo modeling components are to be drawn as nodes in the graph
# illustration, we will now get the node list and the color list, which describes how to color nodes
# according to their communities (which is based on community_map)
if type_of_graph == 'bipartite':
list_of_node_lists = [
list_of_nodes
for list_tuple in numbered_community_map.values()
for list_of_nodes in list_tuple
]
# list_of_node_lists is (as it implies) a list of lists, so we will use the list comprehension
# below to flatten the list and get our one-dimensional node list
node_list = [
node for sublist in list_of_node_lists for node in sublist
]
color_list = []
# Now, we will find the first community that a node appears in and color the node based on that community
# In community_map, certain nodes may appear in multiple communities, and we have chosen to give preference
# to the first community a node appears in
for node in node_list:
not_found = True
for community_key in numbered_community_map:
if not_found and node in (
numbered_community_map[community_key][0] +
numbered_community_map[community_key][1]):
color_list.append(community_key)
not_found = False
# Find top_nodes (one of the two "groups" of nodes in a bipartite graph), which will be used to
# determine the graph layout
if model_graph.number_of_nodes() > 0 and nx.is_connected(
model_graph):
# An index of 1 used because this tends to place constraint nodes on the left, which is
# consistent with the else case
top_nodes = nx.bipartite.sets(model_graph)[1]
else:
top_nodes = {
node
for node in model_graph.nodes()
if node in constraint_variable_map
}
if pos is None: # The case where the user has not provided their own layout
pos = nx.bipartite_layout(model_graph, top_nodes)
else: # This covers the case that type_of_community_map is 'constraint' or 'variable'
# Constraints are in the first list of the tuples in community map and variables are in the second list
position = 0 if type_of_graph == 'constraint' else 1
list_of_node_lists = list(i[position]
for i in numbered_community_map.values())
# list_of_node_lists is (as it implies) a list of lists, so we will use the list comprehension
# below to flatten the list and get our one-dimensional node list
node_list = [
node for sublist in list_of_node_lists for node in sublist
]
# Now, we will find the first community that a node appears in and color the node based on
# that community (in numbered_community_map, certain nodes may appear in multiple communities,
# and we have chosen to give preference to the first community a node appears in)
color_list = []
for node in node_list:
not_found = True
for community_key in numbered_community_map:
if not_found and node in numbered_community_map[
community_key][position]:
color_list.append(community_key)
not_found = False
# Note - there is no strong reason to choose spring layout; it just creates relatively clean graphs
if pos is None: # The case where the user has not provided their own layout
pos = nx.spring_layout(model_graph)
# Define color_map
color_map = plt.cm.get_cmap('viridis', len(numbered_community_map))
# Create the figure and draw the graph
fig = plt.figure()
nx.draw_networkx_nodes(model_graph,
pos,
nodelist=node_list,
node_size=40,
cmap=color_map,
node_color=color_list)
nx.draw_networkx_edges(model_graph, pos, alpha=0.5)
# Make the main title
graph_type = type_of_graph.capitalize()
community_map_type = self.type_of_community_map.capitalize()
main_graph_title = "%s graph - colored using %s community map" % (
graph_type, community_map_type)
main_font_size = 14
plt.suptitle(main_graph_title, fontsize=main_font_size)
# Define a dict that will be used for the graph subtitle
subtitle_naming_dict = {
'bipartite':
'Nodes are variables and constraints & Edges are variables in a constraint',
'constraint': 'Nodes are constraints & Edges are common variables',
'variable': 'Nodes are variables & Edges are shared constraints'
}
# Make the subtitle
subtitle_font_size = 11
plt.title(subtitle_naming_dict[type_of_graph],
fontsize=subtitle_font_size)
if filename is None:
plt.show()
else:
plt.savefig(filename)
plt.close()
# Return the figure and pos, the position dictionary used for the graph layout
return fig, pos
def create_ef_instance(scenario_tree,
ef_instance_name="MASTER",
verbose_output=False,
generate_weighted_cvar=False,
cvar_weight=None,
risk_alpha=None,
cc_indicator_var_name=None,
cc_alpha=0.0):
#
# create the new and empty binding instance.
#
# scenario tree must be "linked" with a set of instances
# to used this function
scenario_instances = {}
for scenario in scenario_tree.scenarios:
if scenario._instance is None:
raise ValueError("Cannot construct extensive form instance. "
"The scenario tree does not appear to be linked "
"to any Pyomo models. Missing model for scenario "
"with name: %s" % (scenario.name))
scenario_instances[scenario.name] = scenario._instance
binding_instance = ConcreteModel(name=ef_instance_name)
root_node = scenario_tree.findRootNode()
opt_sense = minimize \
if (scenario_tree._scenarios[0]._instance_objective.is_minimizing()) \
else maximize
#
# validate cvar options, if specified.
#
cvar_excess_vardatas = []
if generate_weighted_cvar:
if (cvar_weight is None) or (cvar_weight < 0.0):
raise RuntimeError(
"Weight of CVaR term must be >= 0.0 - value supplied=" +
str(cvar_weight))
if (risk_alpha is None) or (risk_alpha <= 0.0) or (risk_alpha >= 1.0):
raise RuntimeError(
"CVaR risk alpha must be between 0 and 1, exclusive - value supplied="
+ str(risk_alpha))
if verbose_output:
print("Writing CVaR weighted objective")
print("CVaR term weight=" + str(cvar_weight))
print("CVaR alpha=" + str(risk_alpha))
print("")
# create the eta and excess variable on a per-scenario basis,
# in addition to the constraint relating to the two.
cvar_eta_variable_name = "CVAR_ETA_" + str(root_node._name)
cvar_eta_variable = Var()
binding_instance.add_component(cvar_eta_variable_name,
cvar_eta_variable)
excess_var_domain = NonNegativeReals if (opt_sense == minimize) else \
NonPositiveReals
compute_excess_constraint = \
binding_instance.COMPUTE_SCENARIO_EXCESS = \
ConstraintList()
for scenario in scenario_tree._scenarios:
cvar_excess_variable_name = "CVAR_EXCESS_" + scenario._name
cvar_excess_variable = Var(domain=excess_var_domain)
binding_instance.add_component(cvar_excess_variable_name,
cvar_excess_variable)
compute_excess_expression = cvar_excess_variable
compute_excess_expression -= scenario._instance_cost_expression
compute_excess_expression += cvar_eta_variable
if opt_sense == maximize:
compute_excess_expression *= -1
compute_excess_constraint.add(
(0.0, compute_excess_expression, None))
cvar_excess_vardatas.append(
(cvar_excess_variable, scenario._probability))
# the individual scenario instances are sub-blocks of the binding instance.
for scenario in scenario_tree._scenarios:
scenario_instance = scenario_instances[scenario._name]
binding_instance.add_component(str(scenario._name), scenario_instance)
# Now deactivate the scenario instance Objective since we are creating
# a new master objective
scenario._instance_objective.deactivate()
# walk the scenario tree - create variables representing the
# common values for all scenarios associated with that node, along
# with equality constraints to enforce non-anticipativity. also
# create expected cost variables for each node, to be computed via
# constraints/objectives defined in a subsequent pass. master
# variables are created for all nodes but those in the last
# stage. expected cost variables are, for no particularly good
# reason other than easy coding, created for nodes in all stages.
if verbose_output:
print("Creating variables for master binding instance")
_cmap = binding_instance.MASTER_CONSTRAINT_MAP = ComponentMap()
for stage in scenario_tree._stages[:-1]: # skip the leaf stage
for tree_node in stage._tree_nodes:
# create the master blending variable and constraints for this node
master_blend_variable_name = \
"MASTER_BLEND_VAR_"+str(tree_node._name)
master_blend_constraint_name = \
"MASTER_BLEND_CONSTRAINT_"+str(tree_node._name)
# don't create master variables for derived
# stage variables as they will not be used in
# the problem, and their values would likely
# never be consistent with what is stored on the
# scenario variables
master_variable_index = Set(
initialize=sorted(tree_node._standard_variable_ids),
ordered=True,
name=master_blend_variable_name + "_index")
binding_instance.add_component(
master_blend_variable_name + "_index", master_variable_index)
master_variable = Var(master_variable_index,
name=master_blend_variable_name)
binding_instance.add_component(master_blend_variable_name,
master_variable)
master_constraint = ConstraintList(
name=master_blend_constraint_name)
binding_instance.add_component(master_blend_constraint_name,
master_constraint)
tree_node_variable_datas = tree_node._variable_datas
for variable_id in sorted(tree_node._standard_variable_ids):
master_vardata = master_variable[variable_id]
vardatas = tree_node_variable_datas[variable_id]
# Don't blend fixed variables
if not tree_node.is_variable_fixed(variable_id):
for scenario_vardata, scenario_probability in vardatas:
_cmap[scenario_vardata] = master_constraint.add(
(master_vardata - scenario_vardata, 0.0))
if generate_weighted_cvar:
cvar_cost_expression_name = "CVAR_COST_" + str(root_node._name)
cvar_cost_expression = Expression(name=cvar_cost_expression_name)
binding_instance.add_component(cvar_cost_expression_name,
cvar_cost_expression)
# create an expression to represent the expected cost at the root node
binding_instance.EF_EXPECTED_COST = \
Expression(initialize=sum(scenario._probability * \
scenario._instance_cost_expression
for scenario in scenario_tree._scenarios))
opt_expression = \
binding_instance.MASTER_OBJECTIVE_EXPRESSION = \
Expression(initialize=binding_instance.EF_EXPECTED_COST)
if generate_weighted_cvar:
cvar_cost_expression_name = "CVAR_COST_" + str(root_node._name)
cvar_cost_expression = \
binding_instance.find_component(cvar_cost_expression_name)
if cvar_weight == 0.0:
# if the cvar weight is 0, then we're only
# doing cvar - no mean.
opt_expression.set_value(cvar_cost_expression)
else:
opt_expression.expr += cvar_weight * cvar_cost_expression
binding_instance.MASTER = Objective(sense=opt_sense, expr=opt_expression)
# CVaR requires the addition of a variable per scenario to
# represent the cost excess, and a constraint to compute the cost
# excess relative to eta.
if generate_weighted_cvar:
# add the constraint to compute the master CVaR variable value. iterate
# over scenario instances to create the expected excess component first.
cvar_cost_expression_name = "CVAR_COST_" + str(root_node._name)
cvar_cost_expression = binding_instance.find_component(
cvar_cost_expression_name)
cvar_eta_variable_name = "CVAR_ETA_" + str(root_node._name)
cvar_eta_variable = binding_instance.find_component(
cvar_eta_variable_name)
cost_expr = 1.0
for scenario_excess_vardata, scenario_probability in cvar_excess_vardatas:
cost_expr += (scenario_probability * scenario_excess_vardata)
cost_expr /= (1.0 - risk_alpha)
cost_expr += cvar_eta_variable
cvar_cost_expression.set_value(cost_expr)
if cc_indicator_var_name is not None:
if verbose_output is True:
print("Creating chance constraint for indicator variable= " +
cc_indicator_var_name)
print("with alpha= " + str(cc_alpha))
if not isVariableNameIndexed(cc_indicator_var_name):
cc_expression = 0 #??????
for scenario in scenario_tree._scenarios:
scenario_instance = scenario_instances[scenario._name]
scenario_probability = scenario._probability
cc_var = scenario_instance.find_component(
cc_indicator_var_name)
cc_expression += scenario_probability * cc_var
def makeCCRule(expression):
def CCrule(model):
return (1.0 - cc_alpha, cc_expression, None)
return CCrule
cc_constraint_name = "cc_" + cc_indicator_var_name
cc_constraint = Constraint(name=cc_constraint_name,
rule=makeCCRule(cc_expression))
binding_instance.add_component(cc_constraint_name, cc_constraint)
else:
print("multiple cc not yet supported.")
variable_name, index_template = extractVariableNameAndIndex(
cc_indicator_var_name)
# verify that the root variable exists and grab it.
# NOTE: we are using whatever scenario happens to laying around... it might be better to use the reference
variable = scenario_instance.find_component(variable_name)
if variable is None:
raise RuntimeError("Unknown variable=" + variable_name +
" referenced as the CC indicator variable.")
# extract all "real", i.e., fully specified, indices matching the index template.
match_indices = extractComponentIndices(variable, index_template)
# there is a possibility that no indices match the input template.
# if so, let the user know about it.
if len(match_indices) == 0:
raise RuntimeError("No indices match template=" +
str(index_template) + " for variable=" +
variable_name)
# add the suffix to all variable values identified.
for index in match_indices:
variable_value = variable[index]
cc_expression = 0 #??????
for scenario in scenario_tree._scenarios:
scenario_instance = scenario_instances[scenario._name]
scenario_probability = scenario._probability
cc_var = scenario_instance.find_component(
variable_name)[index]
cc_expression += scenario_probability * cc_var
def makeCCRule(expression):
def CCrule(model):
return (1.0 - cc_alpha, cc_expression, None)
return CCrule
indexasname = ''
for c in str(index):
if c not in ' ,':
indexasname += c
cc_constraint_name = "cc_" + variable_name + "_" + indexasname
cc_constraint = Constraint(name=cc_constraint_name,
rule=makeCCRule(cc_expression))
binding_instance.add_component(cc_constraint_name,
cc_constraint)
return binding_instance
def _perform_branch_and_bound(solve_data):
solve_data.explored_nodes = 0
root_node = solve_data.working_model
root_util_blk = root_node.GDPopt_utils
config = solve_data.config
# Map unfixed disjunct -> list of deactivated constraints
root_util_blk.disjunct_to_nonlinear_constraints = ComponentMap()
# Map relaxed disjunctions -> list of unfixed disjuncts
root_util_blk.disjunction_to_unfixed_disjuncts = ComponentMap()
# Preprocess the active disjunctions
for disjunction in root_util_blk.disjunction_list:
assert disjunction.active
disjuncts_fixed_True = []
disjuncts_fixed_False = []
unfixed_disjuncts = []
# categorize the disjuncts in the disjunction
for disjunct in disjunction.disjuncts:
if disjunct.indicator_var.fixed:
if disjunct.indicator_var.value == 1:
disjuncts_fixed_True.append(disjunct)
elif disjunct.indicator_var.value == 0:
disjuncts_fixed_False.append(disjunct)
else:
pass # raise error for fractional value?
else:
unfixed_disjuncts.append(disjunct)
# update disjunct lists for predetermined disjunctions
if len(disjuncts_fixed_False) == len(disjunction.disjuncts) - 1:
# all but one disjunct in the disjunction is fixed to False.
# Remaining one must be true. If not already fixed to True, do so.
if not disjuncts_fixed_True:
disjuncts_fixed_True = unfixed_disjuncts
unfixed_disjuncts = []
disjuncts_fixed_True[0].indicator_var.fix(1)
elif disjuncts_fixed_True and disjunction.xor:
assert len(
disjuncts_fixed_True
) == 1, "XOR (only one True) violated: %s" % disjunction.name
disjuncts_fixed_False.extend(unfixed_disjuncts)
unfixed_disjuncts = []
# Make sure disjuncts fixed to False are properly deactivated.
for disjunct in disjuncts_fixed_False:
disjunct.deactivate()
# Deactivate nonlinear constraints in unfixed disjuncts
for disjunct in unfixed_disjuncts:
nonlinear_constraints_in_disjunct = [
constr
for constr in disjunct.component_data_objects(Constraint,
active=True)
if constr.body.polynomial_degree() not in _linear_degrees
]
for constraint in nonlinear_constraints_in_disjunct:
constraint.deactivate()
if nonlinear_constraints_in_disjunct:
# TODO might be worthwhile to log number of nonlinear constraints in each disjunction
# for later branching purposes
root_util_blk.disjunct_to_nonlinear_constraints[
disjunct] = nonlinear_constraints_in_disjunct
root_util_blk.disjunction_to_unfixed_disjuncts[
disjunction] = unfixed_disjuncts
pass
# Add the BigM suffix if it does not already exist. Used later during nonlinear constraint activation.
# TODO is this still necessary?
if not hasattr(root_node, 'BigM'):
root_node.BigM = Suffix()
# Set up the priority queue
queue = solve_data.bb_queue = []
solve_data.created_nodes = 0
unbranched_disjunction_indices = [
i for i, disjunction in enumerate(root_util_blk.disjunction_list)
if disjunction in root_util_blk.disjunction_to_unfixed_disjuncts
]
sort_tuple = BBNodeData(
obj_lb=float('-inf'),
obj_ub=float('inf'),
is_screened=False,
is_evaluated=False,
num_unbranched_disjunctions=len(unbranched_disjunction_indices),
node_count=0,
unbranched_disjunction_indices=unbranched_disjunction_indices,
)
heappush(queue, (sort_tuple, root_node))
# Do the branch and bound
while len(queue) > 0:
# visit the top node on the heap
# from pprint import pprint
# pprint([(
# x[0].node_count, x[0].obj_lb, x[0].obj_ub, x[0].num_unbranched_disjunctions
# ) for x in sorted(queue)])
node_data, node_model = heappop(queue)
config.logger.info("Nodes: %s LB %.10g Unbranched %s" %
(solve_data.explored_nodes, node_data.obj_lb,
node_data.num_unbranched_disjunctions))
# Check time limit
elapsed = get_main_elapsed_time(solve_data.timing)
if elapsed >= config.time_limit:
config.logger.info('GDPopt-LBB unable to converge bounds '
'before time limit of {} seconds. '
'Elapsed: {} seconds'.format(
config.time_limit, elapsed))
no_feasible_soln = float('inf')
solve_data.LB = node_data.obj_lb if solve_data.objective_sense == minimize else -no_feasible_soln
solve_data.UB = no_feasible_soln if solve_data.objective_sense == minimize else -node_data.obj_lb
config.logger.info('Final bound values: LB: {} UB: {}'.format(
solve_data.LB, solve_data.UB))
solve_data.results.solver.termination_condition = tc.maxTimeLimit
return True
# Handle current node
if not node_data.is_screened:
# Node has not been evaluated.
solve_data.explored_nodes += 1
new_node_data = _prescreen_node(node_data, node_model, solve_data)
heappush(
queue,
(new_node_data, node_model)) # replace with updated node data
elif node_data.obj_lb < node_data.obj_ub - config.bound_tolerance and not node_data.is_evaluated:
# Node has not been fully evaluated.
# Note: infeasible and unbounded nodes will skip this condition, because of strict inequality
new_node_data = _evaluate_node(node_data, node_model, solve_data)
heappush(
queue,
(new_node_data, node_model)) # replace with updated node data
elif node_data.num_unbranched_disjunctions == 0 or node_data.obj_lb == float(
'inf'):
# We have reached a leaf node, or the best available node is infeasible.
original_model = solve_data.original_model
copy_var_list_values(
from_list=node_model.GDPopt_utils.variable_list,
to_list=original_model.GDPopt_utils.variable_list,
config=config,
)
solve_data.LB = node_data.obj_lb if solve_data.objective_sense == minimize else -node_data.obj_ub
solve_data.UB = node_data.obj_ub if solve_data.objective_sense == minimize else -node_data.obj_lb
solve_data.master_iteration = solve_data.explored_nodes
if node_data.obj_lb == float('inf'):
solve_data.results.solver.termination_condition = tc.infeasible
elif node_data.obj_ub == float('-inf'):
solve_data.results.solver.termination_condition = tc.unbounded
else:
solve_data.results.solver.termination_condition = tc.optimal
return
else:
_branch_on_node(node_data, node_model, solve_data)
def generate_model_graph(model, type_of_graph, with_objective=True, weighted_graph=True,
use_only_active_components=True):
"""
Creates a networkX graph of nodes and edges based on a Pyomo optimization model
This function takes in a Pyomo optimization model, then creates a graphical representation of the model with
specific features of the graph determined by the user (see Parameters below).
(This function is designed to be called by detect_communities, but can be used solely for the purpose of
creating model graphs as well.)
Parameters
----------
model: Block
a Pyomo model or block to be used for community detection
type_of_graph: str
a string that specifies the type of graph that is created from the model
'constraint' creates a graph based on constraint nodes,
'variable' creates a graph based on variable nodes,
'bipartite' creates a graph based on constraint and variable nodes (bipartite graph).
with_objective: bool, optional
a Boolean argument that specifies whether or not the objective function is included in the graph; the
default is True
weighted_graph: bool, optional
a Boolean argument that specifies whether a weighted or unweighted graph is to be created from the Pyomo
model; the default is True (type_of_graph='bipartite' creates an unweighted graph regardless of this parameter)
use_only_active_components: bool, optional
a Boolean argument that specifies whether inactive constraints/objectives are included in the networkX graph
Returns
-------
bipartite_model_graph/projected_model_graph: nx.Graph
a NetworkX graph with nodes and edges based on the given Pyomo optimization model
number_component_map: dict
a dictionary that (deterministically) maps a number to a component in the model
constraint_variable_map: dict
a dictionary that maps a numbered constraint to a list of (numbered) variables that appear in the constraint
"""
# Start off by making a bipartite graph (regardless of the value of type_of_graph), then if
# type_of_graph = 'variable' or 'constraint', we will "collapse" this bipartite graph into a variable node
# or constraint node graph
# Initialize the data structure needed to keep track of edges in the graph (this graph will be made
# without edge weights, because edge weights are not useful for this bipartite graph)
edge_set = set()
bipartite_model_graph = nx.Graph() # Initialize NetworkX graph for the bipartite graph
constraint_variable_map = {} # Initialize map of the variables in constraint equations
# Make a dict of all the components we need for the NetworkX graph (since we cannot use the components directly
# in the NetworkX graph)
if with_objective:
component_number_map = ComponentMap((component, number) for number, component in enumerate(
model.component_data_objects(ctype=(Constraint, Var, Objective), active=use_only_active_components,
descend_into=True,
sort=SortComponents.deterministic)))
else:
component_number_map = ComponentMap((component, number) for number, component in enumerate(
model.component_data_objects(ctype=(Constraint, Var), active=use_only_active_components, descend_into=True,
sort=SortComponents.deterministic)))
# Create the reverse of component_number_map, which will be used in detect_communities to convert the node numbers
# to their corresponding Pyomo modeling components
number_component_map = dict((number, comp) for comp, number in component_number_map.items())
# Add the components as nodes to the bipartite graph
bipartite_model_graph.add_nodes_from([node_number for node_number in range(len(component_number_map))])
# Loop through all constraints in the Pyomo model to determine what edges need to be created
for model_constraint in model.component_data_objects(ctype=Constraint, active=use_only_active_components,
descend_into=True):
numbered_constraint = component_number_map[model_constraint]
# Create a list of the variable numbers that occur in the given constraint equation
numbered_variables_in_constraint_equation = [component_number_map[constraint_variable]
for constraint_variable in
identify_variables(model_constraint.body)]
# Update constraint_variable_map
constraint_variable_map[numbered_constraint] = numbered_variables_in_constraint_equation
# Create a list of all the edges that need to be created based on the variables in this constraint equation
edges_between_nodes = [(numbered_constraint, numbered_variable_in_constraint)
for numbered_variable_in_constraint in numbered_variables_in_constraint_equation]
# Update edge_set based on the determined edges between nodes
edge_set.update(edges_between_nodes)
# This if statement will be executed if the user chooses to include the objective function as a node in
# the model graph
if with_objective:
# Use a loop to account for the possibility of multiple objective functions
for objective_function in model.component_data_objects(ctype=Objective, active=use_only_active_components,
descend_into=True):
numbered_objective = component_number_map[objective_function]
# Create a list of the variable numbers that occur in the given objective function
numbered_variables_in_objective = [component_number_map[objective_variable]
for objective_variable in identify_variables(objective_function)]
# Update constraint_variable_map
constraint_variable_map[numbered_objective] = numbered_variables_in_objective
# Create a list of all the edges that need to be created based on the variables in the objective function
edges_between_nodes = [(numbered_objective, numbered_variable_in_objective)
for numbered_variable_in_objective in numbered_variables_in_objective]
# Update edge_set based on the determined edges between nodes
edge_set.update(edges_between_nodes)
# Add edges to bipartite_model_graph (the order in which edges are added can affect community detection, so
# sorting prevents any unpredictable changes)
bipartite_model_graph.add_edges_from(sorted(edge_set))
if type_of_graph == 'bipartite': # This is the case where the user wants a bipartite graph, which we made above
# Log important information with the following logger function
_event_log(model, bipartite_model_graph, set(constraint_variable_map), type_of_graph, with_objective)
# Return the bipartite NetworkX graph, the dictionary of node numbers mapped to their respective Pyomo
# components, and the map of constraints to the variables they contain
return bipartite_model_graph, number_component_map, constraint_variable_map
# At this point of the code, we will create the projected version of the bipartite
# model graph (based on the specific value of type_of_graph)
constraint_nodes = set(constraint_variable_map)
if type_of_graph == 'constraint':
graph_nodes = constraint_nodes
else:
variable_nodes = set(number_component_map) - constraint_nodes
graph_nodes = variable_nodes
if weighted_graph:
projected_model_graph = nx.bipartite.weighted_projected_graph(bipartite_model_graph, graph_nodes)
else:
projected_model_graph = nx.bipartite.projected_graph(bipartite_model_graph, graph_nodes)
# Log important information with the following logger function
_event_log(model, projected_model_graph, set(constraint_variable_map), type_of_graph, with_objective)
# Return the projected NetworkX graph, the dictionary of node numbers mapped to their respective Pyomo
# components, and the map of constraints to the variables they contain
return projected_model_graph, number_component_map, constraint_variable_map