desdeo 1.1.3__py3-none-any.whl → 2.0.0__py3-none-any.whl

desdeo/problem/gurobipy_evaluator.py

@@ -0,0 +1,562 @@
+"""Defines an evaluator compatible with the Problem JSON format and transforms it into a GurobipyModel."""
+
+from operator import eq as _eq
+from operator import le as _le
+import warnings
+
+import gurobipy as gp
+import numpy as np
+
+from desdeo.problem.json_parser import FormatEnum, MathParser
+from desdeo.problem.schema import (
+    Constant,
+    Constraint,
+    ConstraintTypeEnum,
+    Objective,
+    Problem,
+    ScalarizationFunction,
+    TensorConstant,
+    TensorVariable,
+    Variable,
+    VariableTypeEnum
+
+)
+
+
+class GurobipyEvaluatorError(Exception):
+    """Raised when an error within the GurobipyEvaluator class is encountered."""
+
+
+class GurobipyEvaluatorWarning(UserWarning):
+    """Raised when the problem contains features that are poorly supported in gurobipy."""
+
+
+class GurobipyEvaluator:
+    """Defines as evaluator that transforms an instance of Problem into a GurobipyModel."""
+
+    # gp.Model does not support these, so the evaluator will handle them
+    objective_functions: dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr]
+    scalarizations: dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr]
+    extra_functions: dict[
+        str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr | gp.GenExpr | int | float
+    ]
+    constants: dict[str, int | float | list[int] | list[float]]
+
+    model: gp.Model
+
+    def __init__(self, problem: Problem):
+        """Initialized the evaluator.
+
+        Args:
+            problem (Problem): the problem to be transformed in a GurobipyModel.
+        """
+        self.model = gp.Model(problem.name)
+        self.objective_functions = {}
+        self.scalarizations = {}
+        self.extra_functions = {}
+        self.constants = {}
+        self.mvars = {}
+
+        # set the parser
+        self.parse = MathParser(to_format=FormatEnum.gurobipy).parse
+
+        # Add variables
+        self.model = self.init_variables(problem)
+
+        # Add constants, if any
+        if problem.constants is not None:
+            self.constants = self.init_constants(problem)
+
+        # Add extra expressions, if any
+        if problem.extra_funcs is not None:
+            self.extra_functions = self.init_extras(problem)
+
+        # Add objective function expressions
+        self.objective_functions = self.init_objectives(problem)
+
+        # Add constraints, if any
+        if problem.constraints is not None:
+            self.model = self.init_constraints(problem)
+
+        # Add scalarization functions, if any
+        if problem.scalarization_funcs is not None:
+            self.scalarizations = self.init_scalarizations(problem)
+
+        self.problem = problem
+
+    def init_variables(self, problem: Problem) -> gp.Model:
+        """Add variables to the GurobipyModel.
+
+        Args:
+            problem (Problem): problem from which to extract the variables.
+
+        Raises:
+            GurobipyEvaluatorError: when a problem in extracting the variables is encountered.
+                I.e., the variables are of a non supported type.
+
+        Returns:
+            GurobipyModel: the GurobipyModel with the variables added as attributes.
+        """
+        for var in problem.variables:
+            if isinstance(var, Variable):
+                # handle regular variables
+                lowerbound = var.lowerbound if var.lowerbound is not None else float("-inf")
+                upperbound = var.upperbound if var.upperbound is not None else float("inf")
+
+                # figure out the variable type
+                match var.variable_type:
+                    case VariableTypeEnum.integer:
+                        # variable is integer
+                        domain = gp.GRB.INTEGER
+                    case VariableTypeEnum.real:
+                        # variable is real
+                        domain = gp.GRB.CONTINUOUS
+                    case VariableTypeEnum.binary:
+                        domain = gp.GRB.BINARY
+                    case _:
+                        msg = f"Could not figure out the type for variable {var}."
+                        raise GurobipyEvaluatorError(msg)
+
+                # add the variable to the model
+                gvar = self.model.addVar(lb=lowerbound, ub=upperbound, vtype=domain, name=var.symbol)
+                # set the initial value, if one has been defined
+                if var.initial_value is not None:
+                    gvar.setAttr("Start", var.initial_value)
+
+            elif isinstance(var, TensorVariable):
+                # handle tensor variables, i.e., vectors etc..
+                lowerbounds = var.get_lowerbound_values() if var.lowerbounds is not None else np.full(var.shape, float("-inf")).tolist()
+                upperbounds = var.get_upperbound_values() if var.upperbounds is not None else np.full(var.shape, float("inf")).tolist()
+
+                # figure out the variable type
+                match var.variable_type:
+                    case VariableTypeEnum.integer:
+                        # variable is integer
+                        domain = gp.GRB.INTEGER
+                    case VariableTypeEnum.real:
+                        # variable is real
+                        domain = gp.GRB.CONTINUOUS
+                    case VariableTypeEnum.binary:
+                        domain = gp.GRB.BINARY
+                    case _:
+                        msg = f"Could not figure out the type for variable {var}."
+                        raise GurobipyEvaluatorError(msg)
+
+                # add the variable to the model
+                gvar = self.model.addMVar(shape=tuple(var.shape), lb=np.array(lowerbounds), ub=np.array(upperbounds), vtype=domain, name=var.symbol)
+                # set the initial value, if one has been defined
+                if var.initial_values is not None:
+                    gvar.setAttr("Start", np.array(var.get_initial_values()))
+                self.mvars[var.symbol] = gvar
+
+
+        # update the model before returning, so that other expressions can reference the variables
+        self.model.update()
+
+        return self.model
+
+    def init_constants(self, problem: Problem) -> dict[str, int | float | list[int] | list[float]]:
+        """Add constants to a GurobipyEvaluator.
+
+        Gurobi does not really have constants, so this function instead
+        stores them in a dict, that is then stored in the evaluator.
+        This is necessary to get the MathParser to understand the constants
+        used in the problem, but updating them at a later point will not update
+        any expression referencing them. The expressions that have been defined
+        using these constants will keep using the numeric value of the constant
+        at the time when the expression was created.
+        Thus, it might be best to avoid using constants if you are intending
+        to use the gurobipy solver.
+
+        Args:
+            problem (Problem): problem from which to extract the constants.
+
+        Raises:
+            GurobipyEvaluatorError: when the domain of a constant cannot be figured out.
+
+        Returns:
+            dict[str, int | float]: a dict containing the constants.
+        """
+        constants: dict[str, int | float | list[int] | list[float]] = {}
+        for con in problem.constants:
+            if isinstance(con, Constant):
+                constants[con.symbol] = con.value
+            elif isinstance(con, TensorConstant):
+                constants[con.symbol] = con.get_values()
+
+        return constants
+
+    def init_extras(
+        self, problem: Problem
+    ) -> dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr | gp.GenExpr | int | float]:
+        """Add extra function expressions to a Gurobipy Model.
+
+        Gurobi does not support extra expressions natively, so this function instead
+        stores them in a dict to be used by the evaluator.
+        This is necessary to get the MathParser to understand the extra expressions
+        used in the problem, but updating them at a later point will not update
+        any expression referencing them. The expressions that have been defined
+        using these extra expressions will keep using the value of the extra expression
+        at the time when the expression was created.
+        Thus, it might be best to avoid using extra expressions if you are intending
+        to use the gurobipy solver.
+
+        Args:
+            problem (Problem): problem from which the extract the extra function expressions.
+
+        Returns:
+            GurobipyModel: the GurobipyModel with the expressions added as attributes.
+        """
+        extra_functions: dict[
+            str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr | gp.GenExpr | int | float
+        ] = {}
+
+        for extra in problem.extra_funcs:
+            extra_functions[extra.symbol] = self.parse(extra.func, callback=self.get_expression_by_name)
+
+        return extra_functions
+
+    def init_objectives(
+        self, problem: Problem
+    ) -> dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr]:
+        """Add objective function expressions to a Gurobipy Model.
+
+        Does not yet add any actual gurobipy optimization objectives, only creates a dict containing the
+        expressions of the objectives. The objective expressions are stored in the
+        evaluator and appropiate gurobipy objective must be added to the model before solving.
+
+        Args:
+            problem (Problem): problem from which to extract the objective function expresions.
+
+        Returns:
+            dict: dict containing the objective functions.
+        """
+        objective_functions: dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr] = {}
+        for obj in problem.objectives:
+            gp_expr = self.parse(obj.func, callback=self.get_expression_by_name)
+            if isinstance(gp_expr, int | float):
+                warnings.warn(
+                    "One or more of the problem objectives seems to be a constant.",
+                    GurobipyEvaluatorWarning,
+                    stacklevel=2,
+                )
+            if isinstance(gp_expr, gp.GenExpr):
+                msg = f"Gurobi does not support objective functions that are not linear or quadratic {gp_expr}"
+                raise GurobipyEvaluatorError(msg)
+
+            objective_functions[obj.symbol] = gp_expr
+
+            # the obj.symbol_min objectives are used when optimizing and building scalarizations etc...
+            objective_functions[f"{obj.symbol}_min"] = -gp_expr if obj.maximize else gp_expr
+
+        return objective_functions
+
+    def init_constraints(self, problem: Problem) -> gp.Model:
+        """Add constraint expressions to a Gurobipy Model.
+
+        Args:
+            problem (Problem): the problem from which to extract the constraint function expressions.
+            model (GurobipyModel): the GurobipyModel to add the exprssions to.
+
+        Raises:
+            GurobipyEvaluatorError: when an unsupported constraint type is encountered.
+
+        Returns:
+            GurobipyModel: the GurobipyModel with the constraint expressions added.
+        """
+        for cons in problem.constraints:
+            gp_expr = self.parse(cons.func, callback=self.get_expression_by_name)
+
+            match con_type := cons.cons_type:
+                case ConstraintTypeEnum.LTE:
+                    # constraints in DESDEO are defined such that they must be less than zero
+                    gp_expr = _le(gp_expr, 0)
+                case ConstraintTypeEnum.EQ:
+                    gp_expr = _eq(gp_expr, 0)
+                case _:
+                    msg = f"Constraint type of {con_type} not supported. Must be one of {ConstraintTypeEnum}."
+                    raise GurobipyEvaluatorError(msg)
+
+            self.model.addConstr(gp_expr, name=cons.symbol)
+
+        self.model.update()
+        return self.model
+
+    def init_scalarizations(
+        self, problem: Problem
+    ) -> dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr]:
+        """Add scalrization expressions to a gurobipy model.
+
+        Scalarizations work identically to objectives, except they are stored in a different
+        dict in the GurobipyModel. If you want to solve the problem using a scalarization, the
+        evaluator needs to set it as an optimization target first.
+
+        Args:
+            problem (Problem): the problem from which to extract the scalarization function expressions.
+
+        Returns:
+            dict: the dict with the scalarization expressions. Scalarization functions are always minimized.
+        """
+        scalarizations: dict[str, gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr] = {}
+
+        for scal in problem.scalarization_funcs:
+            scalarizations[scal.symbol] = self.parse(scal.func, self.get_expression_by_name)
+
+        return scalarizations
+
+    def add_constraint(self, constraint: Constraint) -> gp.Constr:
+        """Add a constraint expression to a GurobipyModel.
+
+        If adding a lot of constraints, this function may end up being very slow compared
+        to adding the constraints to the stored model directly, because of the model.update() calls.
+
+        Args:
+            constraint (Constraint): the constraint function expression.
+
+        Raises:
+            GurobipyEvaluatorError: when an unsupported constraint type is encountered.
+
+        Returns:
+            gurobipy.Constr: The gurobipy constraint that was added.
+        """
+        gp_expr = self.parse(constraint.func, self.get_expression_by_name)
+
+        match con_type := constraint.cons_type:
+            case ConstraintTypeEnum.LTE:
+                # constraints in DESDEO are defined such that they must be less than zero
+                gp_expr = _le(gp_expr, 0)
+            case ConstraintTypeEnum.EQ:
+                gp_expr = _eq(gp_expr, 0)
+            case _:
+                msg = f"Constraint type of {con_type} not supported. Must be one of {ConstraintTypeEnum}."
+                raise GurobipyEvaluatorError(msg)
+
+        return_cons = self.model.addConstr(gp_expr, name=constraint.symbol)
+        self.model.update()
+        return return_cons
+
+    def add_objective(self, obj: Objective):
+        """Adds an objective function expression to a GurobipyModel.
+
+        Does not yet add any actual gurobipy optimization objectives, only adds them to the dict
+        containing the expressions of the objectives. The objective expressions are stored in the
+        GurobipyModel and the evaluator must add the appropiate gurobipy objective before solving.
+
+        Args:
+            obj (Objective): the objective function expression to be added.
+        """
+        gp_expr = self.parse(obj.func, self.get_expression_by_name)
+        if isinstance(gp_expr, int | float):
+            warnings.warn(
+                "One or more of the problem objectives seems to be a constant.", GurobipyEvaluatorWarning, stacklevel=2
+            )
+        if isinstance(gp.GenExpr, int):
+            msg = f"Gurobi does not support objective functions that are not linear or quadratic {gp_expr}"
+            raise GurobipyEvaluatorError(msg)
+
+        self.objective_functions[obj.symbol] = gp_expr
+
+        # the obj.symbol_min objectives are used when optimizing and building scalarizations etc...
+        self.objective_functions[f"{obj.symbol}_min"] = -gp_expr if obj.maximize else gp_expr
+
+    def add_scalarization_function(self, scal: ScalarizationFunction):
+        """Adds a scalrization expression to a gurobipy model.
+
+        Scalarizations work identically to objectives, except they are stored in a different
+        dict in the GurobipyModel. If you want to solve the problem using a scalarization, the
+        evaluator needs to set it as an optimization target first.
+
+        Args:
+            scal (ScalarizationFunction): The scalarization function to be added.
+        """
+        self.scalarizations[scal.symbol] = self.parse(scal.func, self.get_expression_by_name)
+
+    def add_variable(self, var: Variable | TensorVariable) -> gp.Var | gp.MVar:
+        """Add variables to the GurobipyModel.
+
+        If adding a lot of variables, this function may end up being very slow compared
+        to adding the variables to the stored model directly, because of the model.update() calls.
+
+        Args:
+            var (Variable): The definition of the variable to be added.
+
+        Raises:
+            GurobipyEvaluatorError: when a problem in extracting the variables is encountered.
+                I.e., the variables are of a non supported type.
+
+        Returns:
+            gp.Var: the variable that was added to the model.
+        """
+        if isinstance(var, Variable):
+            # handle regular variables
+            lowerbound = var.lowerbound if var.lowerbound is not None else float("-inf")
+            upperbound = var.upperbound if var.upperbound is not None else float("inf")
+
+            # figure out the variable type
+            match var.variable_type:
+                case VariableTypeEnum.integer:
+                    # variable is integer
+                    domain = gp.GRB.INTEGER
+                case VariableTypeEnum.real:
+                    # variable is real
+                    domain = gp.GRB.CONTINUOUS
+                case VariableTypeEnum.binary:
+                    domain = gp.GRB.BINARY
+                case _:
+                    msg = f"Could not figure out the type for variable {var}."
+                    raise GurobipyEvaluatorError(msg)
+
+            # add the variable to the model
+            gvar = self.model.addVar(lb=lowerbound, ub=upperbound, vtype=domain, name=var.symbol)
+            # set the initial value, if one has been defined
+            if var.initial_value is not None:
+                gvar.setAttr("Start", var.initial_value)
+        elif isinstance(var, TensorVariable):
+            # handle tensor variables, i.e., vectors etc..
+            lowerbounds = var.get_lowerbound_values() if var.lowerbounds is not None else np.full(var.shape, float("-inf")).tolist()
+            upperbounds = var.get_upperbound_values() if var.upperbounds is not None else np.full(var.shape, float("inf")).tolist()
+
+            # figure out the variable type
+            match var.variable_type:
+                case VariableTypeEnum.integer:
+                    # variable is integer
+                    domain = gp.GRB.INTEGER
+                case VariableTypeEnum.real:
+                    # variable is real
+                    domain = gp.GRB.CONTINUOUS
+                case VariableTypeEnum.binary:
+                    domain = gp.GRB.BINARY
+                case _:
+                    msg = f"Could not figure out the type for variable {var}."
+                    raise GurobipyEvaluatorError(msg)
+
+            # add the variable to the model
+            gvar = self.model.addMVar(shape=tuple(var.shape), lb=np.array(lowerbounds), ub=np.array(upperbounds), vtype=domain, name=var.symbol)
+            # set the initial value, if one has been defined
+            if var.initial_values is not None:
+                gvar.setAttr("Start", np.array(var.get_initial_values()))
+            self.mvars[var.symbol] = gvar
+
+        self.model.update()
+        return gvar
+
+    def get_expression_by_name(
+        self, name: str
+    ) -> gp.Var | gp.MVar | gp.LinExpr | gp.QuadExpr | gp.MLinExpr | gp.MQuadExpr | gp.GenExpr | int | float:
+        """Returns a gurobipy expression corresponding to the name.
+
+        Only looks for variables, objective functions, scalarizations, extra functions, and constants.
+        This will not find constraints.
+
+        Args:
+            name (str): The symbol of the expression.
+
+        Returns:
+            gurobipy expression: A mathematical expression that gp.Model can use either as a constraint or an objective
+        """
+        expression = self.model.getVarByName(name)
+        if expression is None:
+            # check if an MVar by checking gurobi.MVars stored in the evaluator directly,
+            # which results in terms multiplied by zero being removed from the equations:
+            if name in self.mvars:
+                expression = self.mvars[name]
+            elif name in self.objective_functions:
+                expression = self.objective_functions[name]
+            elif name in self.scalarizations:
+                expression = self.scalarizations[name]
+            elif name in self.extra_functions:
+                expression = self.extra_functions[name]
+            elif name in self.constants:
+                expression = self.constants[name]
+        return expression
+
+
+    def get_values(self) -> dict[str, float | int | bool | list[float] | list[int]]:   # noqa: C901
+        """Get the values from the Gurobipy Model in a dict.
+
+        The keys of the dict will be the symbols defined in the problem utilized to initialize the evaluator.
+
+        Returns:
+            dict[str, float | int | bool]: a dict with keys equivalent to the symbols defined in self.problem.
+        """
+        result_dict = {}
+
+        for var in self.problem.variables:
+            # if var is type MVar, get the values of MVar
+            if var.symbol in self.mvars:
+                result_dict[var.symbol] = self.mvars[var.symbol].getAttr(gp.GRB.Attr.X)
+            else:
+                result_dict[var.symbol] = self.model.getVarByName(var.symbol).getAttr(gp.GRB.Attr.X)
+
+        for obj in self.problem.objectives:
+            result_dict[obj.symbol] = self.objective_functions[obj.symbol].getValue()
+
+        if self.problem.constants is not None:
+            for con in self.problem.constants:
+                result_dict[con.symbol] = self.constants[con.symbol]
+
+        if self.problem.extra_funcs is not None:
+            for extra in self.problem.extra_funcs:
+                result_dict[extra.symbol] = self.extra_functions[extra.symbol].getValue()
+
+        if self.problem.constraints is not None:
+            for const in self.problem.constraints:
+                result_dict[const.symbol] = -self.model.getConstrByName(const.symbol).getAttr("Slack")
+
+        if self.problem.scalarization_funcs is not None:
+            for scal in self.problem.scalarization_funcs:
+                result_dict[scal.symbol] = self.scalarizations[scal.symbol].getValue()
+
+        return result_dict
+
+    def remove_constraint(self, symbol: str):
+        """Removes a constraint from the model.
+
+        If removing a lot of constraints or dealing with a very large model this function
+        may be slow because of the model.update() calls. Accessing the stored model directly
+        may be faster.
+
+        Args:
+            symbol (str): a str representing the symbol of the constraint to be removed.
+        """
+        self.model.remove(self.model.getConstrByName(symbol))
+        self.model.update()
+
+    def remove_variable(self, symbol: str):
+        """Removes a variable from the model.
+
+        If removing a lot of variables or dealing with a very large model this function
+        may be slow because of the model.update() calls. Accessing the stored model directly
+        may be faster.
+
+        Args:
+            symbol (str): a str representing the symbol of the variable to be removed.
+        """
+        if symbol in self.mvars:
+            self.model.remove(self.mvars[symbol])
+            self.mvars.pop(symbol)
+        else:
+            self.model.remove(self.model.getVarByName(symbol))
+        self.model.update()
+
+    def set_optimization_target(self, target: str, maximize: bool = False):  # noqa: FBT001, FBT002
+        """Sets a minimization objective to match the target objective or scalarization of the gurobipy model.
+
+        Args:
+            target (str): an str representing a symbol. Needs to match an objective function or scaralization
+            function already found in the model.
+            maximize (bool): If true, the target function is maximized instead of minimized
+
+        Raises:
+            GurobipyEvaluatorError: the given target was not an attribute of the gurobipy model.
+        """
+        if not ((target in self.objective_functions) or (target in self.scalarizations)):
+            msg = f"The gurobipy model has no objective or scalarization named {target}."
+            raise GurobipyEvaluatorError(msg)
+
+        obj_expr = self.get_expression_by_name(target)
+
+        if maximize:
+            self.model.setObjective(obj_expr, sense=gp.GRB.MAXIMIZE)
+        else:
+            self.model.setObjective(obj_expr)