|gurobi_feasrelax||( model, relaxobjtype, minrelax, penalties, params=NULL )|
This function computes a feasibility relaxation for the input model argument. The feasibility relaxation is a model that, when solved, minimizes the amount by which the solution violates the bounds and linear constraints of the original model. You must provide a penalty to associate with relaxing each individual bound or constraint (through the penalties argument). These penalties are interpreted in different ways, depending on the value of the relaxobjtype argument.
model: The model list must contain a valid Gurobi model. See the model argument section for more information.
relaxobjtype: The approach used to impose penalties on violations.
If you specify relaxobjtype=0, the objective for the feasibility relaxation is to minimize the sum of the weighted magnitudes of the bound and constraint violations.
If you specify relaxobjtype=1, the objective for the feasibility relaxation is to minimize the weighted sum of the squares of the bound and constraint violations.
If you specify relaxobjtype=2, the objective for the feasibility relaxation is to minimize the weighted count of bound and constraint violations.
In all cases, the weights are taken from penalties$lb, penalties$ub and penalties$rhs. You can provide the special penalty value Inf to indicate that the corresponding bound or constraint cannot be relaxed.
minrelax: The minrelax argument is a boolean that controls the type of feasibility relaxation that is created. If minrelax=FALSE, optimizing the returned model gives a solution that minimizes the cost of the violation. If minrelax=TRUE, optimizing the returned model finds a solution that minimizes the original objective, but only from among those solutions that minimize the cost of the violation. Note that gurobi_feasrelax must solve an optimization problem to find the minimum possible relaxation when minrelax=TRUE, which can be quite expensive.
penalties: The penalties argument is a
list of lists, having the following optional named components (default:
lb Penalty for violating each lower bound.
ub Penalty for violating each upper bound.
rhs Penalty for violating each constraint.
To give an example, if a constraint with penalties.rhs value p is violated by 2.0, it would contribute 2*p to the feasibility relaxation objective for relaxobjtype=0, 2*2*p for relaxobjtype=1, and p for relaxobjtype=2.
params: The params list, when provided, contains a list of modified Gurobi parameters. See the params argument section for more information.
A list containing two named components:
result$model, a list variable, as described in the model argument section.
result$feasobj, a scalar. If minrelax==TRUE this is the relaxation problem objective value, 0.0 otherwise.
penalties <- list()
model <- gurobi_read('stein9.mps')
penalties$lb <- rep(1,length(model$lb))
penalties$ub <- rep(1,length(model$ub))
penalties$rhs <- rep(1,length(model$rhs))
feasrelaxresult <- gurobi_feasrelax(model, 0, FALSE, penalties)