Attribute Examples#
Gurobi attribute handling is designed to be orthogonal, meaning that you only need to use a small number of routines to work with a large number of attributes. In particular:
The names and meanings of the various Gurobi attributes remain constant across the different programming language APIs, although some decoration is required in each language.
Given the type of an attribute (double, integer, etc.) and the programming language you wish to use it from, you simply need to identify the appropriate routine for that attribute type in that language in order to query or modify that attribute.
Consider the LB
attribute, which captures the lower bound on a
variable. You would refer to this attribute as follows in the different
Gurobi APIs:
Language |
Attribute |
---|---|
C |
|
C++ |
|
Java |
|
.NET |
|
Python |
|
To query the value of this attribute for an individual variable in the different APIs, you would do the following:
Language |
Attribute Query Example |
---|---|
C |
|
C++ |
|
Java |
|
.NET |
|
Python |
|
Our APIs also include routines for querying attribute values for multiple variables or constraints at once, which is more efficient.
Attributes are referred to using a set of enum
types in C++, Java,
and .NET (one enum for double-valued attributes, one for int-valued
attributes, etc.). In C and Python, the names listed above are simply
constants that take string values. For example, GRB_DBL_ATTR_LB
is
defined in the C layer as:
#define GRB_DBL_ATTR_LB "LB"
In C and Python, you have the option of using the strings directly when
calling attribute methods. If you wish to do so, note that character
case and underscores are ignored. Thus, MIN_COEFF
and MinCoeff
are equivalent.
One important point to note about attributes modification is that it is
done in a lazy fashion. Modifications don’t actually affect the model
until the next request to either update or optimize the model
(GRBupdatemodel
or GRBoptimize
in C).
Refer to the following detailed examples of how to query or modify attributes from our various APIs. You can also browse our Examples to get a better sense of how to use our attribute interface.
Consider the case where you have a Gurobi model m
. You can retrieve
the number of variables in the model by querying the NumVars
model attribute. This is an integer-valued, scalar attribute, so you use
GRBgetintattr
:
int cols;
error = GRBgetintattr(m, GRB_INT_ATTR_NUMVARS, &cols);
You can also use the name of the attribute directly:
int cols;
error = GRBgetintattr(m, "NumVars", &cols);
(Note that attribute capitalization doesn’t matter in the C interface,
so you could also use "numVars"
or "numvars"
).
If you’ve performed optimization on the model, the optimal objective
value can be obtained by querying the ObjVal model attribute.
This is a double-valued, scalar attribute, so you use
GRBgetdblattr
:
double objval;
error = GRBgetdblattr(m, GRB_DBL_ATTR_OBJVAL, &objval);
If you’d like to query the value that a variable takes in the computed
solution, you can query the X variable attribute. This is a
double-valued, vector attribute, so you have a few options for querying
the associated values. You can retrieve the value for a single variable
using GRBgetdblattrelement
:
double x0;
error = GRBgetdblattrelement(m, GRB_DBL_ATTR_X, 0, &x0);
(we query the solution value for variable 0 in this example). You can
also query attribute values for multiple variables using
GRBgetdblattrarray
or GRBgetdblattrlist
:
double x[];
error = GRBgetdblattrarray(m, GRB_DBL_ATTR_X, 0, cols, x);
The former routine retrieves a contiguous set of values (cols
values, starting from index 0
in our example). The latter allows you
to provide a list of indices, and it returns the values for the
corresponding entries.
For each attribute query routine, there’s an analogous set
routine.
To set the upper bound of a variable, for example, you would use
GRBsetdblattrelement
:
error = GRBsetdblattrelement(m, GRB_DBL_ATTR_UB, 0, 0.0);
(In this example, we’ve set the upper bound for variable 0 to 0). You
can set attribute values for multiple variables in a single call using
GRBsetdblattrarray
or GRBsetdblattrlist
.
Consider the case where you have a Gurobi model m
. You can retrieve
the number of variables in the model by querying the NumVars
model attribute using the get
method:
cols = m.get(GRB_IntAttr_NumVars);
If you’ve performed optimization on the model, the optimal objective value can be obtained by querying the ObjVal model attribute:
obj = m.get(GRB_DoubleAttr_ObjVal);
If you’d like to query the value that a variable takes in the computed solution, you can query the X attribute for the corresponding variable object:
vars = m.getVars();
for (int j = 0; j < cols; j++)
xj = vars[j].get(GRB_DoubleAttr_X);
You can also query the value of X
for multiple variables in a single
get
call on the model m
:
double xvals[] = m.get(GRB_DoubleAttr_X, m.GetVars());
For each attribute query method, there’s an analogous set
routine.
To set the upper bound of a variable, for example:
v = m.getVars()[0];
v.set(GRB_DoubleAttr_UB, 0);
(In this example, we’ve set the upper bound for the first variable in the model to 0).
Consider the case where you have a Gurobi model m
. You can retrieve
the number of variables in the model by querying the NumVars
model attribute (which is implemented as a .NET property):
cols = m.NumVars;
If you’ve performed optimization on the model, the optimal objective value can be obtained by querying the ObjVal model attribute:
obj = m.ObjVal;
If you’d like to query the value that a variable takes in the computed solution, you can query the X attribute for the corresponding variable object:
vars = m.GetVars();
for (int j = 0; j < cols; j++)
xj = vars[j].X;
You can also query the value of X
for multiple variables in a single
call using the Get
method on the model
m
:
double[] xvals = m.Get(GRB.DoubleAttr.X, m.GetVars());
For each attribute query method, there’s an analogous Set
routine.
To set the upper bound of a variable, for example:
v = m.GetVars()[0];
v.UB = 0;
(In this example, we’ve set the upper bound for the first variable in the model to 0).
Consider the case where you have a Gurobi model m
. You can retrieve
the number of variables in the model by querying the NumVars
model attribute using the get
method:
cols = m.get(GRB.IntAttr.NumVars);
If you’ve performed optimization on the model, the optimal objective value can be obtained by querying the ObjVal model attribute:
obj = m.get(GRB.DoubleAttr.ObjVal);
If you’d like to query the value that a variable takes in the computed solution, you can query the X attribute for the corresponding variable object:
vars = m.getVars();
for (int j = 0; j < cols; j++)
xj = vars[j].get(GRB.DoubleAttr.X);
You can also query the value of X
for multiple variables in a single
get
call on the model m
:
double[] xvals = m.get(GRB.DoubleAttr.X, m.getVars());
For each attribute query method, there’s an analogous set
routine.
To set the upper bound of a variable, for example:
v = m.getVars()[0];
v.set(GRB.DoubleAttr.UB, 0);
(In this example, we’ve set the upper bound for the first variable in the model to 0).
Consider the case where you have a Gurobi model m
. You can retrieve
the number of variables in the model by querying the NumVars
model attribute:
print(m.numVars)
(Note that attribute capitalization doesn’t matter in the Python
interface, so you could also query m.NumVars
or m.numvars
).
If you’ve performed optimization on the model, the optimal objective value can be obtained by querying the ObjVal model attribute:
print(m.objVal)
If you’d like to query the value that a variable takes in the computed solution, you can query the X attribute for the corresponding variable object:
for v in m.getVars():
print(v.x)
You can also query the value of X
for multiple variables in a single
getAttr
call on the model m
:
print(m.getAttr(GRB.Attr.X, m.getVars()))
For each attribute query method, there’s an analogous set
routine.
To set the upper bound of a variable, for example:
v = m.getVars()[0]
v.ub = 0
(In this example, we’ve set the upper bound for the first variable in the model to 0).
Consider the case where you have a Gurobi model m
. You can retrieve
the number of variables in the model by querying the NumVars
model attribute (which is implemented as a .NET property):
cols = m.NumVars;
If you’ve performed optimization on the model, the optimal objective value can be obtained by querying the ObjVal model attribute:
obj = m.ObjVal;
If you’d like to query the value that a variable takes in the computed solution, you can query the X attribute for the corresponding variable object:
vars = m.GetVars();
For j As Integer = 0 To cols - 1
xj = vars[j].X;
You can also query the value of X
for multiple variables in a single
call using the Get
method on the model
m
:
xvals = m.Get(GRB.DoubleAttr.X, m.GetVars());
For each attribute query method, there’s an analogous Set
routine.
To set the upper bound of a variable, for example:
v = m.GetVars()[0];
v.UB = 0;
(In this example, we’ve set the upper bound for the first variable in the model to 0).