from __future__ import annotations
import inspect
import logging
import os
import threading
from collections.abc import Sequence
from enum import Enum
from pathlib import Path
from typing import TYPE_CHECKING, Any, ClassVar
from gams.core.gdx import GMS_UEL_IDENT_SIZE
import gamspy as gp
import gamspy._algebra.expression as expression
import gamspy._algebra.operation as operation
import gamspy._miro as miro
import gamspy._symbols.implicits as implicits
import gamspy._validation as validation
import gamspy.utils as utils
from gamspy._backend.backend import backend_factory
from gamspy._convert import (
GamsConverter,
LatexConverter,
get_convert_solver_options,
)
from gamspy._model_instance import ModelInstance
from gamspy._options import (
EXECUTION_OPTIONS,
MODEL_ATTR_OPTION_MAP,
ConvertOptions,
FreezeOptions,
Options,
)
from gamspy.exceptions import GamspyException, ValidationError
if TYPE_CHECKING:
from typing import Literal, TextIO
import pandas as pd
from gamspy import Container, Equation, Parameter, Variable
from gamspy._algebra.expression import Expression
from gamspy._algebra.operation import Operation
from gamspy._backend.engine import EngineClient
from gamspy._backend.neos import NeosClient
from gamspy._symbols.implicits import ImplicitParameter, ImplicitVariable
from gamspy._symbols.symbol import Symbol
from gamspy.math import MathOp
GMS_MAX_LINE_LENGTH = 80000
MAX_MODEL_DECLARATION_LENGTH = 75
MAX_SYMBOL_NAME_LENGTH = GMS_UEL_IDENT_SIZE - 1
MAX_MATCHING_LENGTH = (
MAX_SYMBOL_NAME_LENGTH + MAX_SYMBOL_NAME_LENGTH + MAX_MODEL_DECLARATION_LENGTH + 1
)
MAX_NUM_MODEL_ELEMS = int(GMS_MAX_LINE_LENGTH / MAX_MATCHING_LENGTH)
IS_MIRO_INIT = os.getenv("MIRO", False)
logger = logging.getLogger("MODEL")
logger.setLevel(logging.INFO)
stream_handler = logging.StreamHandler()
stream_handler.setLevel(logging.INFO)
formatter = logging.Formatter("[%(name)s - %(levelname)s] %(message)s")
stream_handler.setFormatter(formatter)
logger.addHandler(stream_handler)
[docs]
class Problem(Enum):
"""An enumeration for problem all problem types"""
LP = "LP"
"Linear Programming"
NLP = "NLP"
"""Non-Linear Programming"""
QCP = "QCP"
"""Quadratically Constrained Programs"""
DNLP = "DNLP"
"""Nonlinear Programming with Discontinuous Derivatives"""
MIP = "MIP"
"""Mixed Integer Programming"""
RMIP = "RMIP"
"""Relaxed Mixed Integer Program"""
MINLP = "MINLP"
"""Mixed Integer Nonlinear Program"""
RMINLP = "RMINLP"
"""Relaxed Mixed Integer Nonlinear Program"""
MIQCP = "MIQCP"
"""Mixed Integer Quadratically Constrained Program"""
RMIQCP = "RMIQCP"
"""Relaxed Mixed Integer Quadratically Constrained Program"""
MCP = "MCP"
"""Mixed Complementarity Problem"""
CNS = "CNS"
"""Constrained Nonlinear System"""
MPEC = "MPEC"
"""Mathematical Programs with Equilibrium Constraints"""
RMPEC = "RMPEC"
"""Relaxed Mathematical Program with Equilibrium Constraints"""
EMP = "EMP"
"""Extended Mathematical Program"""
@classmethod
def values(cls):
"""Convenience function to return all values of enum"""
return list(cls._value2member_map_.keys())
def __str__(self) -> str:
return self.value
INT_TO_PROBLEM: dict[int, str] = {
1: str(Problem.LP),
2: str(Problem.MIP),
3: str(Problem.RMIP),
4: str(Problem.NLP),
5: str(Problem.MCP),
6: str(Problem.MPEC),
7: str(Problem.RMPEC),
8: str(Problem.CNS),
9: str(Problem.DNLP),
10: str(Problem.RMINLP),
11: str(Problem.MINLP),
12: str(Problem.QCP),
13: str(Problem.MIQCP),
14: str(Problem.RMIQCP),
15: str(Problem.EMP),
}
[docs]
class Sense(Enum):
"""An enumeration for sense types"""
MIN = "MIN"
"""Minimize the objective."""
MAX = "MAX"
"""Maximize the objective."""
FEASIBILITY = "FEASIBILITY"
"""Assess feasibility."""
@classmethod
def values(cls):
"""Convenience function to return all values of enum"""
return list(cls._value2member_map_.keys())
def __str__(self) -> str:
return self.value
[docs]
class ModelStatus(Enum):
"""An enumeration for model status types"""
OptimalGlobal = 1
"""The solution is optimal, that is, it is feasible (within tolerances) and it has been proven that no other feasible solution with better objective value exists."""
OptimalLocal = 2
"""A local optimum for an NLP has been found. That is, a solution that is feasible (within tolerances) and it has been proven that there exists a neighborhood of this solution in which no other feasible solution with better objective value exists."""
Unbounded = 3
"""The solution is unbounded. This message is reliable if the problem is linear, but occasionally it appears for difficult nonlinear problems that are not truly unbounded, but that lack some strategically placed bounds to limit the variables to sensible values."""
InfeasibleGlobal = 4
"""The problem has been proven to be infeasible. If this was not intended, something is probably misspecified in the logic or the data."""
InfeasibleLocal = 5
"""No feasible point could be found for the NLP problem from the given starting point. It does not necessarily mean that no feasible point exists."""
InfeasibleIntermed = 6
"""The current solution is not feasible, but the solver stopped, either because of a limit (for example, iteration or resource) or because of some sort of difficulty. The solver status will give more information."""
Feasible = 7
"""A feasible solution to a problem without discrete variables has been found."""
Integer = 8
"""A feasible solution to a problem with discrete variables has been found."""
NonIntegerIntermed = 9
"""An incomplete solution to a problem with discrete variables. A feasible solution has not yet been found."""
IntegerInfeasible = 10
"""It has been proven that there is no feasible solution to a problem with discrete variables."""
LicenseError = 11
"""The solver cannot find the appropriate license key needed to use a specific subsolver."""
ErrorUnknown = 12
"""After a solver error the model status is unknown."""
ErrorNoSolution = 13
"""An error occurred and no solution has been returned. No solution will be returned to GAMS because of errors in the solution process."""
NoSolutionReturned = 14
"""A solution is not expected for this solve. For example, the CONVERT solver only reformats the model but does not give a solution."""
SolvedUnique = 15
"""Indicates the solution returned is unique, i.e. no other solution exists. Used for CNS models. Examples where this status could be returned include non-singular linear models, triangular models with constant non-zero elements on the diagonal, and triangular models where the functions are monotone in the variable on the diagonal."""
Solved = 16
"""Indicates the model has been solved: used for CNS models. The solution might or might not be unique. If the solver uses status SOLVED SINGULAR wherever possible then this status implies that the Jacobian is non-singular, i.e. that the solution is at least locally unique."""
SolvedSingular = 17
"""Indicates the CNS model has been solved, but the Jacobian is singular at the solution. This can indicate that other solutions exist, either along a line (for linear models) or a curve (for nonlinear models) including the solution returned."""
UnboundedNoSolution = 18
"""The model is unbounded and no solution can be provided."""
InfeasibleNoSolution = 19
"""The model is infeasible and no solution can be provided."""
[docs]
class SolveStatus(Enum):
"""An enumeration for solve status types"""
NormalCompletion = 1
"""The solver terminated in a normal way."""
IterationInterrupt = 2
"""The solver was interrupted because it used too many iterations. The option `iteration_limit` may be used to increase the iteration limit if everything seems normal."""
ResourceInterrupt = 3
"""The solver was interrupted because it used too much time. The option `time_limit` may be used to increase the time limit if everything seems normal."""
TerminatedBySolver = 4
"""The solver encountered some difficulty and was unable to continue."""
EvaluationInterrupt = 5
"""Too many evaluations of nonlinear terms at undefined values. We recommend to use variable bounds to prevent forbidden operations, such as division by zero. The rows in which the errors occur are listed just before the solution."""
CapabilityError = 6
"""The solver does not have the capability required by the model. For example, some solvers do not support certain types of discrete variables or support a more limited set of functions than other solvers."""
LicenseError = 7
"""The solver cannot find the appropriate license key needed to use a specific subsolver."""
UserInterrupt = 8
"""The user has sent a signal to interrupt the solver."""
SetupError = 9
"""The solver encountered a fatal failure during problem set-up time."""
SolverError = 10
"""The solver encountered a fatal error."""
InternalError = 11
"""The solver encountered an internal fatal error."""
Skipped = 12
"""The entire solve step has been skipped."""
SystemError = 13
"""This indicates a completely unknown or unexpected error condition."""
INTERRUPT_STATUS = [
SolveStatus.IterationInterrupt,
SolveStatus.ResourceInterrupt,
SolveStatus.EvaluationInterrupt,
SolveStatus.UserInterrupt,
SolveStatus.TerminatedBySolver,
]
ERROR_STATUS = {
SolveStatus.CapabilityError: "The solver does not have the capability required by the model.",
SolveStatus.LicenseError: "The solver cannot find the appropriate license key needed to use a specific subsolver.",
SolveStatus.SetupError: "The solver encountered a fatal failure during problem set-up time.",
SolveStatus.SolverError: "The solver encountered a fatal error.",
SolveStatus.InternalError: "The solver encountered an internal fatal error.",
SolveStatus.SystemError: "This indicates a completely unknown or unexpected error condition.",
}
# GAMS name -> GAMSPy name
ATTRIBUTE_MAP = {
"domUsd": "_num_domain_violations",
"etAlg": "_algorithm_time",
"etSolve": "_total_solve_time",
"etSolver": "_total_solver_time",
"iterUsd": "_num_iterations",
"marginals": "_marginals",
"maxInfes": "_max_infeasibility",
"meanInfes": "_mean_infeasibility",
"modelStat": "_status",
"nodUsd": "_num_nodes_used",
"number": "_solve_number",
"numDepnd": "_num_dependencies",
"numDVar": "_num_discrete_variables",
"numEqu": "_num_equations",
"numInfes": "_num_infeasibilities",
"numNLIns": "_num_nonlinear_insts",
"numNLNZ": "_num_nonlinear_zeros",
"numNOpt": "_num_nonoptimalities",
"numNZ": "_num_nonzeros",
"numRedef": "_num_mcp_redefinitions",
"numVar": "_num_variables",
"numVarProj": "_num_bound_projections",
"objEst": "_objective_estimation",
"objVal": "_objective_value",
"procUsed": "_used_model_type",
"resGen": "_model_generation_time",
"resUsd": "_solve_model_time",
"solveStat": "_solve_status",
"sumInfes": "_sum_infeasibilities",
"sysVer": "_solver_version",
}
[docs]
class Model:
"""
Represents a collection of equations and variables to be solved as a mathematical optimization problem.
https://gamspy.readthedocs.io/en/latest/user/basics/model.html
Parameters
----------
container : Container
The Container object that this model belongs to.
name : str, optional
Name of the model. If omitted, a unique name is generated automatically.
description : str, optional
A human-readable description of the model.
equations : Sequence[Equation], optional
A list or sequence of Equation objects that define the constraints of the model.
No equations are used by default.
problem : Problem | str, optional
The type of mathematical problem to solve (e.g., 'LP', 'MIP', 'NLP').
Default is Problem.MIP.
sense : Sense | str, optional
The optimization sense: "MIN" (minimize), "MAX" (maximize), or "FEASIBILITY".
Default is Sense.FEASIBILITY.
objective : Variable | Expression | Operation, optional
The objective to optimize. Can be a scalar variable, an expression, or an operation (like Sum).
matches : dict[Equation | Sequence[Equation], Variable | Sequence[Variable]], optional
Used for defining complementarity problems (MCP). Maps equations to their complementary variables.
limited_variables : Sequence[ImplicitVariable], optional
Allows limiting the domain of variables included in the model to a specific subset.
external_module : str, optional
The name of an external module file (e.g., 'my_external.dll' or 'my_external.so')
where external equations are implemented.
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> i = gp.Set(m, description="items")
>>> p = gp.Parameter(m, description="profits", domain=i)
>>> w = gp.Parameter(m, description="weights", domain=i)
>>> c = gp.Parameter(m, description="capacity")
>>> x = gp.Variable(m, domain=i, type=gp.VariableType.BINARY)
>>> capacity_restriction = gp.Equation(m, definition=gp.Sum(i, w[i] * x[i]) <= c)
>>> # Instantiate the Model
>>> knapsack = gp.Model(
... m,
... equations=m.getEquations(), # Automatically grabs all equations in the container
... problem=gp.Problem.MIP, # Mixed Integer Program
... sense=gp.Sense.MAX, # Maximize profit
... objective=gp.Sum(i, p[i] * x[i]),
... )
"""
# Prefix for auto-generated symbols
_generate_prefix = "autogenerated_"
_autogen_symbols: ClassVar[set[str]] = set()
def __init__(
self,
container: Container | None = None,
name: str | None = None,
description: str = "",
problem: Problem | str = Problem.MIP,
equations: Sequence[Equation] | None = None,
sense: Sense | str = Sense.FEASIBILITY,
objective: Variable | Expression | Operation | None = None,
matches: dict[
Equation | Sequence[Equation],
Variable | Sequence[Variable],
]
| None = None,
limited_variables: Sequence[ImplicitVariable] | None = None,
external_module: str | None = None,
):
self._is_mpsge = False
self._auto_id = "m" + utils._get_unique_name()
if equations is None:
equations = []
self.description = description
if container is not None:
self.container = container
else:
self.container = gp._ctx_managers[(os.getpid(), threading.get_native_id())]
assert self.container is not None
if name is not None:
name = validation.validate_name(name)
self.name = validation.validate_model_name(name)
else:
self.name = self.container._get_symbol_name(prefix="m")
self._matches = matches
self.problem, self.sense = validation.validate_model(
equations, matches, problem, sense
)
self.equations = list(equations)
self._objective = objective
self._objective_variable = self._set_objective_variable(objective)
if (
self._objective_variable is not None
and self._objective_variable.type.lower() != "free"
):
raise ValidationError(
f"Objective variable `{self._objective_variable}` must be a free variable"
)
if limited_variables is not None and not isinstance(
limited_variables, Sequence
):
raise ValidationError(
f"`limited_variables must be an Iterable of ImplicitVariable objects but found {limited_variables}`"
)
self._limited_variables: Sequence[ImplicitVariable] | None = limited_variables
if not self.equations and not self._matches:
raise ValidationError("Model requires at least one equation.")
self._external_module_file: str | None = None
self._external_module: str | None = None
if external_module is not None:
self.external_module = external_module
else:
# To avoid adding it twice
self.container._add_statement(self)
# allow freezing
self._is_frozen: bool = False
# Attributes
self._num_domain_violations: float | None = None
self._algorithm_time: float | None = None
self._total_solve_time: float | None = None
self._total_solver_time: float | None = None
self._num_iterations: float | None = None
self._marginals: float | None = None
self._max_infeasibility: float | None = None
self._mean_infeasibility: float | None = None
self._status: ModelStatus | None = None
self._num_nodes_used: float | None = None
self._solve_number: float | None = None
self._num_dependencies: float | None = None
self._num_discrete_variables: float | None = None
self._num_equations: float | None = None
self._num_infeasibilities: float | None = None
self._num_nonlinear_insts: float | None = None
self._num_nonlinear_zeros: float | None = None
self._num_nonoptimalities: float | None = None
self._num_nonzeros: float | None = None
self._num_mcp_redefinitions: float | None = None
self._num_variables: float | None = None
self._num_bound_projections: float | None = None
self._objective_estimation: float | None = None
self._objective_value: float | None = None
self._used_model_type: str | None = None
self._model_generation_time: float | None = None
self._solve_model_time: float | None = None
self._sum_infeasibilities: float | None = None
self._solve_status: SolveStatus | None = None
self._solver_version: float | None = None
self.container.models.update({self.name: self})
self.container._synch_with_gams()
def _serialize(self) -> dict:
info: dict[str, Any] = {
"name": self.name,
"problem": str(self.problem),
"sense": str(self.sense),
}
# equations
equations = [equation.name for equation in self.equations]
info["equations"] = equations
# matches
if self._matches is not None:
matches: dict = {}
for key, value in self._matches.items():
if isinstance(key, gp.Equation):
if isinstance(value, gp.Variable):
matches[key.name] = value.name
else:
matches[key.name] = (variable.name for variable in value)
else:
assert isinstance(value, gp.Variable)
for equation in key:
matches[equation.name] = value.name
info["_matches"] = matches
# objective variable
if self._objective_variable is not None:
info["_objective_variable"] = self._objective_variable.name
# attributes
for attr_name in ATTRIBUTE_MAP.values():
if attr_name in ("_status", "_solve_status"):
attr_value = getattr(self, attr_name, None)
if attr_value is not None:
attr_value = attr_value.value
info[attr_name] = attr_value
else:
info[attr_name] = getattr(self, attr_name, None)
return info
def __repr__(self) -> str:
return f"Model(name='{self.name}', problem='{self.problem!s}', equations={self.equations}, sense='{self.sense!s}', objective={self._objective_variable}, matches={self._matches}, limited_variables={self._limited_variables}"
def __str__(self) -> str:
return (
f"Model {self.name}:\n Problem Type: {self.problem}\n Sense:"
f" {self.sense}\n Equations: {self.equations}"
)
@property
def num_domain_violations(self) -> float | None:
"""
Number of domain violations in the solution.
Returns
-------
float | None
"""
return self._num_domain_violations
@property
def algorithm_time(self) -> float | None:
"""
Solver dependent timing information. This attribute was intended to allow
solvers to return the elapsed time used by the solve algorithm without
including any model generation, communication, or setup time. However,
solvers are free to adapt this convention and return time-related information
(but not necessarily elapsed time) for executing the solve algorithm. Please
inspect your solver manual for the actual meaning of the value returned in
this attribute.
Returns
-------
float | None
"""
return self._algorithm_time
@property
def total_solve_time(self) -> float | None:
"""
Elapsed time it took to execute a solve statement in total. This model
attribute returns the elapsed time it took to execute a solve statement in total.
This time includes the model generation time, the time to read and write files,
the time to create the solution report and the time taken by the actual solve.
The time is expressed in seconds of wall-clock time.
Returns
-------
float | None
"""
return self._total_solve_time
@property
def total_solver_time(self) -> float | None:
"""
Elapsed time taken by the solver only. This model attribute returns the elapsed
time taken by the solver only. This does not include the GAMS model generation time
and the time taken to report and load the solution back into the GAMS database.
The time is expressed in seconds of wall-clock time.
Returns
-------
float | None
"""
return self._total_solver_time
@property
def num_iterations(self) -> float | None:
"""
Number of iterations used.
Returns
-------
float | None
"""
return self._num_iterations
@property
def marginals(self) -> float | None:
"""
Indicates whether there are marginals.
Returns
-------
float | None
"""
return self._marginals
@property
def max_infeasibility(self) -> float | None:
"""
Maximum of infeasibilities
Returns
-------
float | None
"""
return self._max_infeasibility
@property
def mean_infeasibility(self) -> float | None:
"""
Mean of infeasibilities
Returns
-------
float | None
"""
return self._mean_infeasibility
@property
def status(self) -> ModelStatus | None:
"""
Model status after solve.
Returns
-------
ModelStatus | None
"""
return self._status
@property
def num_nodes_used(self) -> float | None:
"""
Number of nodes used by the MIP solver.
Returns
-------
float | None
"""
return self._num_nodes_used
@property
def solve_number(self) -> float | None:
"""
Number of the last solve.
Returns
-------
float | None
"""
return self._solve_number
@property
def num_dependencies(self) -> float | None:
"""
Number of dependencies in a CNS model.
Returns
-------
float | None
"""
return self._num_dependencies
@property
def num_discrete_variables(self) -> float | None:
"""
Number of discrete variables.
Returns
-------
float | None
"""
return self._num_discrete_variables
@property
def num_equations(self) -> float | None:
"""
Number of equations.
Returns
-------
float | None
"""
return self._num_equations
@property
def num_infeasibilities(self) -> float | None:
"""
Number of infeasibilities.
Returns
-------
float | None
"""
return self._num_infeasibilities
@property
def num_nonlinear_insts(self) -> float | None:
"""
Number of nonlinear instructions.
Returns
-------
float | None
"""
return self._num_nonlinear_insts
@property
def num_nonlinear_zeros(self) -> float | None:
"""
Number of nonlinear nonzeros.
Returns
-------
float | None
"""
return self._num_nonlinear_zeros
@property
def num_nonoptimalities(self) -> float | None:
"""
Number of nonoptimalities.
Returns
-------
float | None
"""
return self._num_nonoptimalities
@property
def num_nonzeros(self) -> float | None:
"""
Number of nonzero entries in the model coefficient matrix.
Returns
-------
float | None
"""
return self._num_nonzeros
@property
def num_mcp_redefinitions(self) -> float | None:
"""
Number of MCP redefinitions.
Returns
-------
float | None
"""
return self._num_mcp_redefinitions
@property
def num_variables(self) -> float | None:
"""
Number of variables.
Returns
-------
float | None
"""
return self._num_variables
@property
def num_bound_projections(self) -> float | None:
"""
Number of bound projections during model generation.
Returns
-------
float | None
"""
return self._num_bound_projections
@property
def objective_estimation(self) -> float | None:
"""
Estimate of the best possible solution for a mixed-integer model
Returns
-------
float | None
"""
return self._objective_estimation
@property
def objective_value(self) -> float | None:
"""
Objective function value
Returns
-------
float | None
"""
return self._objective_value
@property
def used_model_type(self) -> str | None:
"""
Integer number that indicates the used model type.
Returns
-------
str | None
"""
return self._used_model_type
@property
def model_generation_time(self) -> float | None:
"""
Time GAMS took to generate the model in wall-clock seconds.
Returns
-------
float | None
"""
return self._model_generation_time
@property
def solve_model_time(self) -> float | None:
"""
Time the solver used to solve the model in seconds
Returns
-------
float | None
"""
return self._solve_model_time
@property
def solve_status(self) -> SolveStatus | None:
"""
Indicates the solver termination condition.
Returns
-------
SolveStatus | None
"""
return self._solve_status
@property
def sum_infeasibilities(self) -> float | None:
"""
Sum of infeasibilities.
Returns
-------
float | None
"""
return self._sum_infeasibilities
@property
def solver_version(self) -> float | None:
"""
Solver version.
Returns
-------
float | None
"""
return self._solver_version
@property
def external_module(self) -> str | None:
"""
Name of the external module in which the external equations are implemented.
By default, this parameter is set to None. When provided, it triggers the
opening of the specified file using a File statement and incorporates the
file into the model by adding it as an external module.
This feature requires a solid understanding of programming, compilation,
and linking processes. For more information, please refer to the
https://gamspy.readthedocs.io/en/latest/user/advanced/external_equations.html .
Returns
-------
str | None
"""
return self._external_module
@external_module.setter
def external_module(self, value: str | None):
if self._external_module_file is not None:
self.container._add_statement(f"putclose {self._external_module_file};")
write_model_statement = value != self._external_module
self._external_module = None
self._external_module_file = None
if value is not None:
filename = "f" + utils._get_unique_name()
self._external_module_file = filename
self._external_module = value
self.container._add_statement(f"File {filename} / '{value}' /;")
if write_model_statement:
self.container._add_statement(self)
def _generate_obj_var_and_equation(self) -> tuple[Variable, Equation]:
variable = gp.Variable._constructor_bypass(
self.container,
f"{self.name}_objective_variable",
domain=[],
)
equation = gp.Equation._constructor_bypass(
self.container,
f"{self.name}_objective",
domain=[],
)
self._autogen_symbols.add(variable.name)
self._autogen_symbols.add(equation.name)
return variable, equation
def _set_objective_variable(
self,
assignment: None | Variable | Operation | Expression | MathOp = None,
) -> Variable | None:
"""
Returns objective variable. If the assignment is an Expression
or an Operation (Sum, Product etc.), it automatically generates
an objective variable and a equation.
Returns
-------
Variable | None
Raises
------
TypeError
In case assignment is not a Variable, Expression or an Operation.
"""
if assignment is not None and not isinstance(
assignment,
(
gp.Variable,
expression.Expression,
operation.Operation,
gp.math.MathOp,
),
):
raise TypeError(
"Objective must be a Variable or an Expression but"
f" {type(assignment)} given"
)
if self.sense == gp.Sense.FEASIBILITY:
if assignment is not None:
raise ValidationError(
"Cannot set an objective when the sense is FEASIBILITY!"
)
if self.problem in (Problem.CNS, Problem.MCP, Problem.EMP):
return None
variable, equation = self._generate_obj_var_and_equation()
statement = expression.Expression(
implicits.ImplicitEquation(
equation,
name=equation.name,
type=equation.type,
domain=[],
),
"..",
variable == 0,
)
self.container._add_statement(statement)
equation._definition = statement
equation.modified = False
variable.modified = False
if equation.name not in (symbol.name for symbol in self.equations):
self.equations.append(equation)
return variable
if isinstance(
assignment,
(expression.Expression, operation.Operation, gp.math.MathOp),
):
variable, equation = self._generate_obj_var_and_equation()
# Sum((i,j),c[i,j]*x[i,j])->Sum((i,j),c[i,j]*x[i,j]) =e= var
assignment = assignment == variable
# equation .. Sum((i,j),c[i,j]*x[i,j]) =e= var
statement = expression.Expression(
implicits.ImplicitEquation(
equation,
name=equation.name,
type=equation.type,
domain=[],
),
"..",
assignment,
)
self.container._add_statement(statement)
equation._definition = statement
equation.modified = False
variable.modified = False
if equation.name not in (symbol.name for symbol in self.equations):
self.equations.append(equation)
return variable
return assignment
def _generate_solve_string(self) -> str:
solve_statement = [f"solve {self.name} using {self.problem}"]
if self.sense == gp.Sense.FEASIBILITY:
# Set sense as min or max for feasibility
self.sense = gp.Sense("MIN")
if self.problem not in (Problem.MCP, Problem.CNS, Problem.EMP):
solve_statement.append(str(self.sense))
if self._objective_variable is not None:
solve_statement.append(self._objective_variable.gamsRepr())
return " ".join(solve_statement)
def _add_runtime_options(self, options: Options, backend: str = "local") -> None:
for key, value in options.model_dump(exclude_none=True).items():
if key in MODEL_ATTR_OPTION_MAP:
if isinstance(value, bool):
value = int(value)
elif isinstance(value, str):
value = f"'{value}'"
self.container._add_statement(
f"{self.name}.{MODEL_ATTR_OPTION_MAP[key]} = {value};\n"
)
elif key in EXECUTION_OPTIONS:
if key == "loadpoint":
if isinstance(value, os.PathLike):
value = os.path.abspath(value)
if backend == "engine":
value = os.path.relpath(value, self.container.working_directory)
self.container._add_statement(f"{EXECUTION_OPTIONS[key]} '{value}';\n")
def _create_model_attributes(self) -> None:
self.container._add_statement("$offListing")
for attr_name in ATTRIBUTE_MAP:
symbol_name = f"{self._generate_prefix}{attr_name}_{self._auto_id}"
Symbol = gp.Parameter._constructor_bypass(self.container, symbol_name)
Symbol.modified = False
self.container._add_statement(f"{symbol_name} = {self.name}.{attr_name};")
self.container._add_statement("$onListing")
def _update_model_attributes(self) -> None:
gdx_handle = utils._open_gdx_file(
self.container.system_directory, self.container._gdx_out
)
for gams_attr, python_attr in ATTRIBUTE_MAP.items():
symbol_name = f"{self._generate_prefix}{gams_attr}_{self._auto_id}"
data = utils._get_scalar_data(
self.container._gams2np, gdx_handle, symbol_name
)
if python_attr == "_status":
setattr(self, python_attr, ModelStatus(data))
elif python_attr == "_solve_status":
status = SolveStatus(data)
setattr(self, python_attr, status)
if status in INTERRUPT_STATUS:
logger.warning(
f"The solve was interrupted! Solve status: {status.name}. "
"For further information, see https://gamspy.readthedocs.io/en/latest/reference/gamspy._model.html#gamspy.SolveStatus."
)
elif status in ERROR_STATUS:
raise GamspyException(
f"Solve status: {status.name}. {ERROR_STATUS[status]}",
status.value,
)
elif python_attr == "_used_model_type":
setattr(self, python_attr, INT_TO_PROBLEM[int(data)])
else:
setattr(self, python_attr, data)
utils._close_gdx_handle(gdx_handle)
[docs]
def computeInfeasibilities(self) -> dict[str, pd.DataFrame]:
"""
Computes infeasabilities for all equations of the model
Returns
-------
dict[str, pd.DataFrame]
Dictionary of infeasibilities where equation names are keys and
infeasibilities are values
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> i = gp.Set(m, name="i", records=["i1", "i2"])
>>> j = gp.Set(m, name="j", records=["j1", "j2", "j3"])
>>> a = gp.Parameter(m, name="a", domain=i, records=[("i1", 350), ("i2", 600)])
>>> b = gp.Parameter(m, name="b", domain=j, records=[("j1", 400), ("j2", 450), ("j3", 420)])
>>> x = gp.Variable(m, name="x", domain=[i,j], type="Positive")
>>> s = gp.Equation(m, name="s", domain=i)
>>> d = gp.Equation(m, name="d", domain=j)
>>> s[i] = gp.Sum(j, x[i, j]) <= a[i]
>>> d[j] = gp.Sum(i, x[i, j]) >= b[j]
>>> my_model = gp.Model(m, name="my_model", equations=m.getEquations(), problem="LP", sense="min", objective=gp.Sum((i, j), x[i, j]))
>>> summary = my_model.solve()
>>> infeasibilities = my_model.computeInfeasibilities()
>>> infeasibilities["s"].infeasibility.item()
320.0
"""
infeas_dict = {}
for equation in self.equations:
if equation.records is None:
continue
infeas_rows = utils._calculate_infeasibilities(equation)
infeas_dict[equation.name] = infeas_rows
if equation._definition is not None:
names = equation._definition._find_all_symbols()
names = [
name
for name in names
if name in self.container.data
and isinstance(self.container[name], gp.Variable)
]
for name in names:
variable = self.container[name]
infeas_rows = utils._calculate_infeasibilities(variable)
infeas_dict[variable.name] = infeas_rows
return infeas_dict
[docs]
def convert(
self,
path: str | Path,
file_format: FileFormat | Sequence[FileFormat],
options: ConvertOptions | None = None,
) -> None:
"""
Converts the model to one or more specified file formats.
Parameters
----------
path : str | Path
Path to the directory where the converted model files will be saved.
file_format : FileFormat | Sequence[FileFormat]
File format(s) to convert the model to. Can be a single FileFormat or a list of FileFormats.
options : ConvertOptions, optional
Additional options to customize the conversion process.
Raises
------
ValueError
If the specified file format is not supported.
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> v = gp.Variable(m, "v")
>>> e = gp.Equation(m, "e", definition= v == 5)
>>> my_model = gp.Model(m, "my_model", problem="LP", equations=[e])
Convert the model into scalar GAMS format.
>>> my_model.convert("tmp", gp.FileFormat.GAMS) # doctest: +SKIP
Add conversion options
>>> options = gp.ConvertOptions(GDXNames=False)
>>> my_model.convert("jacobian", file_format=gp.FileFormat.GDXJacobian, options=options) # doctest: +SKIP
"""
path = Path(path)
os.makedirs(path, exist_ok=True)
solver_options = get_convert_solver_options(path, file_format, options)
self.solve(solver="convert", solver_options=solver_options)
[docs]
def getEquationListing(
self,
infeasibility_threshold: float | None = None,
) -> str:
"""
Returns the generated equations.
Parameters
----------
infeasibility_threshold: float, optional
Filters out equations with infeasibilities that are above this value.
Returns
-------
str
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> i = gp.Set(m, records=["item1", "item2"])
>>> v = gp.Variable(m, domain=i)
>>> z = gp.Variable(m)
>>> e = gp.Equation(m, domain=i)
>>> e[i] = v[i] * z >= 5
>>> e2 = gp.Equation(m, domain=i)
>>> e2[i] = v[i] - z >= 10
>>> model = gp.Model(m, "test", equations=[e, e2], problem="NLP", sense="MIN", objective=z)
>>> summary = model.solve(options=gp.Options(equation_listing_limit=10))
>>> print(model.getEquationListing())
e(item1).. (0)*v(item1) + (0)*z =G= 5 ; (LHS = 0, INFES = 5 ****)
e(item2).. (0)*v(item2) + (0)*z =G= 5 ; (LHS = 0, INFES = 5 ****)
e2(item1).. v(item1) - z =G= 10 ; (LHS = 0, INFES = 10 ****)
e2(item2).. v(item2) - z =G= 10 ; (LHS = 0, INFES = 10 ****)
"""
listings = []
for equation in self.equations:
listing = equation.getEquationListing(
infeasibility_threshold=infeasibility_threshold
)
if listing:
listings.append(listing)
return "\n".join(listings)
[docs]
def getVariableListing(self) -> str:
"""
Returns the variable listing.
Returns
-------
str
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> i = gp.Set(m, records=["item1", "item2"])
>>> v = gp.Variable(m, domain=i)
>>> z = gp.Variable(m)
>>> e = gp.Equation(m, domain=i)
>>> e[i] = v[i] * z >= 5
>>> model = gp.Model(m, "test", equations=[e], problem="NLP", sense="MIN", objective=z)
>>> summary = model.solve(options=gp.Options(variable_listing_limit=10))
>>> print(model.getVariableListing())
v(item1)
(.LO, .L, .UP, .M = -INF, 0, +INF, 0)
(0) e(item1)
<BLANKLINE>
v(item2)
(.LO, .L, .UP, .M = -INF, 0, +INF, 0)
(0) e(item2)
<BLANKLINE>
z
(.LO, .L, .UP, .M = -INF, 0, +INF, 0)
(0) e(item1)
(0) e(item2)
<BLANKLINE>
"""
if not hasattr(self, "_variables"):
raise ValidationError(
"The model must be solved with `variable_listing_limit` option for this functionality to work."
)
listings = []
for variable in self._variables:
listings.append(variable.getVariableListing())
return "\n".join(listings)
[docs]
def interrupt(self) -> None:
"""
Sends interrupt signal to the running job.
Examples
--------
>>> import gamspy as gp
>>> import threading
>>> m = gp.Container()
>>> # ... define model ...
>>> model = gp.Model(m, "my_model", problem="LP", equations=[])
>>> # In a separate thread or signal handler:
>>> model.interrupt() # doctest: +SKIP
"""
self.container._interrupt()
[docs]
def freeze(
self,
modifiables: list[Parameter | ImplicitParameter],
options: Options | None = None,
) -> None:
"""
Instantiates a model instance. After calling freeze, only modifiables can be modified.
Parameters
----------
modifiables : list[Parameter | ImplicitParameter]
Modifiable symbols.
options : Options | None, optional
GAMSPy options, by default None
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> p = gp.Parameter(m, "p", records=1)
>>> v = gp.Variable(m, "v")
>>> e = gp.Equation(m, "e", definition=v >= p)
>>> model = gp.Model(m, "test", equations=[e], problem="LP", sense="MIN", objective=v)
>>> model.freeze(modifiables=[p])
>>> p.setRecords(2) # Modify parameter without regenerating the model
>>> summary = model.solve()
>>> model.unfreeze()
"""
self._is_frozen = True
if options is None:
options = self.container._options
self.instance = ModelInstance(
self.container, self, modifiables, options, self.container.output
)
[docs]
def unfreeze(self) -> None:
"""
Unfreezes the model
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> p = gp.Parameter(m, "p", records=1)
>>> v = gp.Variable(m, "v")
>>> e = gp.Equation(m, "e", definition=v >= p)
>>> model = gp.Model(m, "test", equations=[e], problem="LP", sense="MIN", objective=v)
>>> model.freeze(modifiables=[p])
>>> p.setRecords(2) # Modify parameter without regenerating the model
>>> summary = model.solve()
>>> model.unfreeze()
"""
self._is_frozen = False
self.instance.close_license_session()
[docs]
def solve(
self,
solver: str | None = None,
options: Options | None = None,
solver_options: dict | str | Path | None = None,
freeze_options: FreezeOptions | None = None,
output: TextIO | None = None,
backend: Literal["local", "engine", "neos"] = "local",
client: EngineClient | NeosClient | None = None,
load_symbols: list[Symbol] | None = None,
) -> pd.DataFrame | None:
"""
Solves the model with given options.
https://gamspy.readthedocs.io/en/latest/user/basics/model.html
Parameters
----------
solver : str, optional
Solver name
options : Options, optional
GAMSPy options.
solver_options : dict | str | Path, optional
Dictionary of solver options or path to an existing option file.
freeze_options : FreezeOptions, optional
Options to solve a frozen model.
output : TextIO, optional
Output redirection target. Set this to sys.stdout to see the output in your terminal.
backend : str, optional
Backend to run on
client : EngineClient, NeosClient, optional
EngineClient to communicate with GAMS Engine or NEOS Client to communicate with NEOS Server
load_symbols : list[Symbol], optional
Specifies the symbols that need to be loaded. If not given, all symbols are loaded after solve.
Returns
-------
DataFrame, optional
Summary of the solve
Raises
------
ValidationError
In case engine_config is not provided for `engine` backend or
neos_client is not provided for `neos` backend.
ValueError
In case problem is not in possible problem types
ValueError
In case sense is different than "MIN" or "MAX"
Examples
--------
>>> import sys
>>> import os
>>> import gamspy as gp
>>> m = gp.Container()
>>> i = gp.Set(m, description="items")
>>> p = gp.Parameter(m, description="profits", domain=i)
>>> w = gp.Parameter(m, description="weights", domain=i)
>>> c = gp.Parameter(m, description="capacity")
>>> x = gp.Variable(m, domain=i, type=gp.VariableType.BINARY)
>>> capacity_restriction = gp.Equation(m, definition=gp.Sum(i, w[i] * x[i]) <= c)
>>> knapsack = gp.Model(
... m,
... equations=m.getEquations(), # Automatically grabs all equations in the container
... problem=gp.Problem.MIP, # Mixed Integer Program
... sense=gp.Sense.MAX, # Maximize profit
... objective=gp.Sum(i, p[i] * x[i]),
... )
Basic usage
>>> knapsack.solve() # doctest: +SKIP
Change solver
>>> knapsack.solve(solver="CPLEX") # doctest: +SKIP
Redirect output
>>> knapsack.solve(output=sys.stdout) # doctest: +SKIP
Add generic solve options
>>> knapsack.solve(options=gp.Options(iteration_limit=2)) # doctest: +SKIP
Add solver-specific options
>>> knapsack.solve(solver="CPLEX", solver_options={"preind": "off"}) # doctest: +SKIP
Solve on your own machine
>>> knapsack.solve() # doctest: +SKIP
Solve on GAMS Engine
>>> client = gp.EngineClient(
... host=os.getenv("ENGINE_URL", "https://<host_link>"),
... username=os.getenv("ENGINE_USER"),
... password=os.getenv("ENGINE_PASSWORD"),
... namespace=os.getenv("ENGINE_NAMESPACE"),
... )
>>> knapsack.solve(backend="engine", client=client) # doctest: +SKIP
Solve on NEOS Server
>>> client = gp.NeosClient(
... email=os.getenv("NEOS_EMAIL"),
... username=os.getenv("NEOS_USER"),
... password=os.getenv("NEOS_PASSWORD"),
... )
>>> knapsack.solve(backend="neos", client=client) # doctest: +SKIP
"""
if output is None:
output = self.container.output
if solver is None:
solver = utils.getDefaultSolvers(self.container.system_directory)[
str(self.problem).upper()
].lower()
else:
if not isinstance(solver, str):
raise TypeError(
f"`solver` argument must be of type `str` but given `{type(solver)}`"
)
solver = solver.lower()
if solver == "conopt":
solver = "conopt4"
validation.validate_solver_args(
self.container.system_directory,
backend,
solver,
self.problem,
options,
output,
load_symbols,
)
validation.validate_equations(self)
if options is None:
options = self.container._options
if isinstance(solver_options, (str, Path)):
solver_options = Path(solver_options)
# Only for local until GAMS Engine and NEOS Server backends adopt the new GP_SolveLine option.
if backend == "local":
frame = inspect.currentframe().f_back
assert isinstance(options, gp.Options)
options._frame = frame
if solver_options is not None:
self.container.writeSolverOptions(solver, solver_options)
if self._is_frozen:
instance_options = FreezeOptions()
if freeze_options is not None:
instance_options = freeze_options
summary = self.instance.solve(
solver, instance_options, solver_options, output
)
return summary
self.container._add_statement(self.getDeclaration())
self._add_runtime_options(options, backend)
self.container._add_statement(self._generate_solve_string() + "\n")
self._create_model_attributes()
options._set_model_info(solver, self.problem, solver_options)
runner = backend_factory(
self.container,
options,
solver,
solver_options,
output,
backend,
client,
self,
load_symbols,
)
summary = runner.run(relaxed_domain_mapping=False, gams_to_gamspy=True)
if IS_MIRO_INIT:
miro._write_default_gdx_miro(self.container)
return summary
[docs]
def getDeclaration(self) -> str:
"""
Declaration of the Model in GAMS
Returns
-------
str
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> v = gp.Variable(m, "v")
>>> e = gp.Equation(m, "e", definition= v == 5)
>>> my_model = gp.Model(m, "my_model", problem="LP", equations=[e])
>>> my_model.getDeclaration()
'Model my_model / e,my_model_objective /;'
"""
def build_str_with_new_lines(elements: list[str]) -> str:
num_elements = len(elements)
if num_elements < MAX_NUM_MODEL_ELEMS:
return ",".join(elements)
count = 0
return_str = ""
while count < num_elements:
batch = elements[count : count + MAX_NUM_MODEL_ELEMS]
return_str += ",".join(batch) + "\n"
count += MAX_NUM_MODEL_ELEMS
return return_str
equations_in_matches = []
if self._matches is not None:
for key in self._matches:
if isinstance(key, gp.Equation):
equations_in_matches.append(key)
else:
equations_in_matches += list(key)
equations = []
for equation in self.equations:
if self._matches is not None:
if equation not in equations_in_matches:
equations.append(equation.name)
else:
equations.append(equation.name)
equations_str = build_str_with_new_lines(equations)
if self._matches is not None:
matches = []
for key, value in self._matches.items():
if isinstance(key, gp.Equation):
if isinstance(value, gp.Variable):
matches.append(f"{key.name}.{value.name}")
else:
matches.append(
f"{key.name}:({'|'.join([variable.name for variable in value])})"
)
else:
assert isinstance(value, gp.Variable)
matches.append(
f"({'|'.join([equation.name for equation in key])}):{value.name}"
)
matches_str = build_str_with_new_lines(matches)
equations_str = (
",".join([equations_str, matches_str]) if equations else matches_str
)
if self._external_module_file is not None:
equations_str = ",".join([equations_str, self._external_module_file])
if self._limited_variables is not None:
limited_variables_str = ",".join(
[variable.gamsRepr() for variable in self._limited_variables]
)
equations_str += "," + limited_variables_str
model_str = f"Model {self.name}"
if self.description:
model_str += ' "' + self.description + '"'
if equations_str != "":
model_str += f" / {equations_str} /"
model_str += ";"
return model_str
[docs]
def toGams(
self,
path: str | Path,
options: Options | None = None,
*,
dump_gams_state: bool = False,
) -> None:
"""
Generates GAMS model under path/<model_name>.gms.
Parameters
----------
path : str | Path
Path to the directory which will contain the GAMS model.
options : Options | None, optional
GAMSPy options, by default None
dump_gams_state : bool, optional
Whether to dump the state as a GAMS save file, by default False
Raises
------
ValidationError
In case the given options is not of type gp.Options.
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> v = gp.Variable(m, "v")
>>> e = gp.Equation(m, "e", definition= v == 5)
>>> my_model = gp.Model(m, "my_model", problem="LP", equations=[e])
>>> my_model.toGams("tmp") # doctest: +SKIP
================================================================================
GAMS (.gms) file has been generated under ...
================================================================================
"""
if options is not None and not isinstance(options, gp.Options):
raise ValidationError(
f"`options` must be of type gp.Options of found {type(options)}"
)
path = Path(path)
converter = GamsConverter(self, path, options, dump_gams_state)
converter.convert()
[docs]
def toLatex(
self,
path: str | Path,
rename: dict[str, str] | None = None,
*,
generate_pdf: bool = False,
) -> None:
"""
Generates a latex file that contains the model definition under path/<model_name>.tex
Parameters
----------
path : str | Path
Path to the directory which will contain the .tex file.
rename: dict[str, str], optional
A dictionary to rename symbols in the LaTeX file. Keys are GAMSPy symbol names
and values are the names that will be used in the LaTeX file.
generate_pdf: bool, False by default
Generates a pdf file if it is set. Requires pdflatex to be installed.
Examples
--------
>>> import gamspy as gp
>>> m = gp.Container()
>>> v = gp.Variable(m, "v")
>>> e = gp.Equation(m, "e", definition= v == 5)
>>> my_model = gp.Model(m, "my_model", problem="LP", equations=[e])
>>> my_model.toLatex("tmp") # doctest: +SKIP
================================================================================
LaTeX (.tex) file has been generated under ...
================================================================================
"""
path = Path(path)
converter = LatexConverter(self, path, rename)
converter.convert()
if generate_pdf:
converter.to_pdf()
