Parameter¶

class Parameter¶

The Parameter object stores the information regarding a parameter in the algebraic model. It can contain multiple instances, typically numbers (or strings in case of symbolic paramters) or can be scalar. Accessing numerical data of a Parameter can via achieved via the single instances (throgh Parameter.get) or via the Parameter.getValues which returns a DataFrame object containing all the information stored in the parameter, including the indexing. Data can be assigned using AMPL.setData and a DataFrame or using Parameter.set or Parameter.setValues.

get(key)¶

classmethod Parameter.get(key)¶

Syntax

v = get(key)

Description

v = get(key) returns the object corresponding to the specified key. Note that the same effect can be achieved using MATLAB’s indexing operator (round brackets).

Input Arguments

The key of the parameter instance to look up.

Output Arguments

v: The specified parameter instance

Example

Create an indexed variable, get an instance for each, and fix them to some value

ampl.eval('var x{1..5}; var y{1..2, 1..2};');
x = ampl.getVariable('x');
y = ampl.getVariable('y');
x2 = x.get(2);      % or x2 = x(2);
y2_2 = y.get(2,2);  % or y2_2 = y(2,2);
x2.fix(3);
y2_2.fix(4.4);
x.getValues
y.getValues

gives:

ans =
  i1   |  val
  1.0  |  0.0
  2.0  |  3.0
  3.0  |  0.0
  4.0  |  0.0
  5.0  |  0.0

  ans =
  i1   i2   |  val
  1.0  1.0  |  0.0
  1.0  2.0  |  0.0
  2.0  1.0  |  0.0
  2.0  2.0  |  4.4

getValues¶

classmethod Parameter.getValues()¶

Syntax

df = getValues()

Description

df = getValues() returns a dataframe with all the values stored in the parameter, decoupled from the underlying AMPL. To gain access to the data, see DataFrame.

Output Arguments

df: A dataframe containing all the values of the parameter and all its indices

Example

Create a parameter and get its values in a dataframe. Prove that it is disconnected from the underlying AMPL instance:

ampl.eval('param b{i in 1..2} := i;');
b = ampl.getParameter('b');
df = b.getValues();
ampl.close();
ampl
df

gives:

ampl =
AMPL is not running

df =
i in 1 .. 2  |  value
1            |  1.0
2            |  2.0

hasDefault¶

classmethod Parameter.hasDefault()¶

Syntax

v = hasDefault

Description

v = hasDefault returns true if the parameter has a default value set

Output Arguments

v: 1 if the parameter has a default value, 0 otherwise

Example

Create two parameters, one with default and one not and check which one has a default:

ampl.eval('param a; param b{1..2} default 1;');
a = ampl.getParameter('a');
b = ampl.getParameter('b');
a.hasDefault
b.hasDefault

gives:

ans =
  0

ans =
  1

isSymbolic¶

classmethod Parameter.isSymbolic()¶

Syntax

v = isSymbolic

Description

v = isSymbolic returns true if the parameter is symbolic (can store strings rather than numbers)

Output Arguments

v: 1 if the parameter is symbolic, 0 otherwise

Example

Create two parameters and check which one is symbolic:

ampl.eval('param a symbolic; param b{1..2} default 1;');
a = ampl.getParameter('a');
b = ampl.getParameter('b');
a.isSymbolic
b.isSymbolic

gives:

ans =
  1

ans =
  0

set(index, value)¶

classmethod Parameter.set()¶

Syntax

set(value)

set(stringvalue)

set(index, value)

set(index, stringvalue)

Description

Function to set a single value of a parameter.

set(value) and set(stringvalue) set the a scalar parameter to the specified value.

set(index, value) and set(index, stringvalue) set the a scalar parameter to the specified value.

Input Arguments

index: The index of the value to be set
value: The numerical value to set
stringvalue: The literal value to set

Example

Create a few parameters and assign single values to their instances.

ampl.eval('param a{1..5} default 2; param b{1..2, 1..2} symbolic default "default value";');
a = ampl.getParameter('a');
b = ampl.getParameter('b');
ad = a.getInstances
bd = b.getInstances
a.set(2, 42);
b.set([1,2], 'New Value');
 ad = a.getInstances
bd = b.getInstances

gives:

ad =
  2
  2
  2
  2
  2

bd =
 'default value'
 'default value'
 'default value'
 'default value'

ad =
  2
 42
  2
  2
  2

bd =
 'default value'
 'New Value'
 'default value'
 'default value'

setValues¶

classmethod Parameter.setValues()¶

Syntax

setValues(indices, values)

setValues(indices, stringvalues)

setValues(valuesmatrix)

setValues(values)

setValues(stringvalues)

Description

Function to set multiple values of a parameter which has no value assigned. To set data to multiple entities at once, see AMPL.setData.

setValues(indices, values) Sets the values of the parameter at the specified indices to the values specified in values

setValues(indices, stringvalues) Sets the values of the symbolic parameter at the specified indices to the literal values specified in stringvalues

setValues(valuesmatrix) Sets the values of the parameter to the values specified in the matrix values. Valid only for 2-d parameters

setValues(values) Assign the specified double values to the parameter, assigning them to the parameter in the same order as the indices in the entity

setValues(stringvalues) Assign the specified string values to the parameter, assigning them to the parameter in the same order as the indices in the entity

Input Arguments

indices: An array or cell array of indices. Each tuple must be a row in the array.
values: The numerical values to set. Each value must be a row in the array..
stringvalues: The literal values to set. Each value must be a row in the array..
valuesmatrix: The matrix of values to set to a 2-d parameter. For Example, the $$m \times n$$ matrix: $A = \left( \begin{array}{cccc} a_{11} & a_{12} & ... & a_{1n} \\ x_{21} & x_{22} & ... & x_{2n} \\ ... & ... & ... & ... \\ a_{m1} & a_{m2} & ... & a_{mn} \end{array} \right)$ can be assigned to the AMPL parameter: param a{1..m, 1..n};with the statement setValues(A)

Example

Ordered. Create a few parameters and assign specific data specifying the indices of the instances.

ampl.eval('param a{1..5} default 2; param b{1..2, 1..2};');
a = ampl.getParameter('a');
b = ampl.getParameter('b');
indices = [2; 4]; values = [11; 22];
a.setValues(indices, values);
a.getValues

gives:

and for multiple indices, simply:

indices = [1 1, 2,2];
b.setValues(indices, values);
b.getValues

gives:

Bi-dimensional (assigning from matrix)

  ampl.eval('param 2d{1..3, 1..2};');
  values = eye(3,2);
  d = ampl.getParameter('2d');
  d.setValues(values);
  d.getValues

gives::

  ans =
   i1   i2   |  val
   1.0  1.0  |  1.0
   1.0  2.0  |  0.0
   2.0  1.0  |  0.0
   2.0  2.0  |  1.0
   3.0  1.0  |  0.0
   3.0  2.0  |  0.0

Not specifying the indices, it is possible to do operations like:

  ampl.eval('param a{i in 1..5} := i; param b{1..5} default 0;');
  a = ampl.getParameter('a');
  b = ampl.getParameter('b');
  values = a.getInstances;
  b.setValues(values);
  b.display

which assign all the values of a to b, as shown by the output::

  ans =
  i1   |  val
  1.0  |  1.0
  2.0  |  2.0
  3.0  |  3.0
  4.0  |  4.0
  5.0  |  5.0

ObjectiveInstance

Set