Optimization Action Set

Simple Quadratic Program

This section contains PROC CAS code.

Note: Input data must be accessible in your CAS session, either as a CAS table or as a transient-scope table. A CAS table has a two-level name: the first level is your CAS engine libref, and the second level is the table name. You refer to this table in the CAS procedure by specifying only the second level. For more information about two-level names, see Chapter 3, Shared Concepts (SAS Optimization: Mathematical Optimization Procedures). A transient-scope table is called directly from the action and exists in memory for the duration of the action. For more information about accessing data, see SAS Viya: System Programming Guide. For more information about PROC CAS and programming in CASL, see SAS Cloud Analytic Services: CASL Programmer’s Guide and SAS Cloud Analytic Services: CASL Reference.

The following example illustrates how you can use the solveQp action to solve quadratic programs. Suppose you want to solve the following problem:

StartLayout 1st Row 1st Column min 2nd Column 2 x 1 3rd Column plus 4th Column 3 x 2 5th Column plus 6th Column x 1 squared 7th Column plus 8th Column 10 x 2 squared 9th Column plus 10th Column 2.5 x 1 x 2 2nd Row 1st Column subject to 2nd Column x 1 3rd Column minus 4th Column x 2 5th Column less-than-or-equal-to 6th Column 1 7th Column Blank 8th Column Blank 9th Column Blank 10th Column Blank 3rd Row 1st Column Blank 2nd Column x 1 3rd Column plus 4th Column 2 x 2 5th Column greater-than-or-equal-to 6th Column 100 7th Column Blank 8th Column Blank 9th Column Blank 10th Column Blank 4th Row 1st Column Blank 2nd Column x 1 3rd Column Blank 4th Column Blank 5th Column greater-than-or-equal-to 6th Column 0 7th Column Blank 8th Column Blank 9th Column Blank 10th Column Blank 5th Row 1st Column Blank 2nd Column Blank 3rd Column Blank 4th Column x 2 5th Column greater-than-or-equal-to 6th Column 0 7th Column Blank 8th Column Blank 9th Column Blank 10th Column Blank EndLayout

The preceding problem can be expressed in the following manner in an input data table that uses a quadratic programming system (QPS) format. The DATA step assumes that your CAS engine libref is named mycas, but you can substitute any appropriately defined CAS engine libref.

data mycas.example;
   input _id_ field1 $ field2 $ field3 $ field4 field5 $ field6 @;
   datalines;
1  NAME     .      EXAMPLE    .             .         .
2  ROWS     .      .          .             .         .
3  N        OBJ    .          .             .         .
4  L        R1     .          .             .         .
5  G        R2     .          .             .         .
6  COLUMNS  .      .          .             .         .
7  .        X1     R1         1.0           R2        1.0
8  .        X1     OBJ        2.0           .         .
9  .        X2     R1        -1.0           R2        2.0
10 .        X2     OBJ        3.0           .         .
11 RHS      .      .          .             .         .
12 .        RHS    R1         1.0           .         .
13 .        RHS    R2         100           .         .
14 RANGES   .      .          .             .         .
15 BOUNDS   .      .          .             .         .
16 QUADOBJ  .      .          .             .         .
17 .        X1     X1         2.0           .         .
18 .        X1     X2         2.5           .         .
19 .        X2     X2         20            .         .
20 ENDATA   .      .          .             .         .
;

You can use the following call to PROC CAS to solve the QP problem:

proc cas;
   loadactionset "optimization";
   action optimization.solveQp result=r status=s /
      data={name="example"}
      primalOut={name="expout" replace=true}
      dualOut={name="exdout" replace=true}
      numThreads=4
      objSense="min"
      logFreq=1;
   run;
   print r.SolutionSummary; run;
   action table.fetch / table = "expout"; run;
quit;

The primal solution is displayed in Output 2.14.1.

Output 2.14.1: Example 1: Primal Solution Output

Results from table.fetch

Selected Rows from Table EXPOUT
_Index_Objective Function IDRHS IDVariable NameVariable
Type
Linear Objective
Coefficient
Lower BoundUpper BoundVariable ValueVariable
Status
1OBJRHSX1N201.797693E30833.999999914O
2OBJRHSX2N301.797693E30833.000000079O


The progress of the solution is printed to the log, as shown in Output 2.14.2.

Output 2.14.2: Log: Solution Progress

 
NOTE: Active Session now MYSESS.                                                
NOTE: Added action set 'optimization'.                                          
NOTE: The problem EXAMPLE has 2 variables (0 free, 0 fixed).                    
NOTE: The problem has 2 constraints (1 LE, 0 EQ, 1 GE, 0 range).                
NOTE: The problem has 4 constraint coefficients.                                
NOTE: The objective function has 2 Hessian diagonal elements and 1 Hessian      
      elements above the diagonal.                                              
NOTE: The QP presolver value AUTOMATIC is applied.                              
NOTE: The QP presolver removed 0 variables and 0 constraints.                   
NOTE: The QP presolver removed 0 constraint coefficients.                       
NOTE: The presolved problem has 2 variables, 2 constraints, and 4 constraint    
      coefficients.                                                             
NOTE: The QP solver is called.                                                  
NOTE: The Interior Point algorithm is used.                                     
NOTE: The deterministic parallel mode is enabled.                               
NOTE: The Interior Point algorithm is using up to 4 threads.                    
                                        Primal       Bound        Dual          
      Iter  Complement Duality Gap      Infeas      Infeas      Infeas   Time   
         0  4.4604E+03  2.6380E-01  1.7962E-02  1.8143E+00  2.7770E-14      0   
         1  1.2367E+02  7.8255E-03  1.7962E-04  1.8143E-02  1.9285E-14      0   
         2  1.2365E+00  7.8496E-05  1.7973E-06  1.8154E-04  4.6816E-14      0   
         3  1.2364E-02  7.8498E-07  1.7973E-08  1.8154E-06  1.9865E-14      0   
         4  0.0000E+00  7.8498E-09  1.7861E-09  1.8154E-08  3.4973E-07      0   
NOTE: Optimal.                                                                  
NOTE: Objective = 15018.000046.                                                 
NOTE: The Interior Point solve time is 0.00 seconds.                            
 
 
 
 
 


The PRINT statement that follows immediately after the PROC CAS step prints the value of the SolutionSummary results table to the log, as shown in Output 2.14.3.

Output 2.14.3: Log: SolutionSummary Results Table

SolutionSummary: Results from optimization.solveQp

Solution Summary
SolverQP
AlgorithmInterior Point
Objective FunctionOBJ
Solution StatusOptimal
Objective Value15018.000046
  
Primal Infeasibility0
Dual Infeasibility0
Bound Infeasibility0
Duality Gap7.8497853E-9
Complementarity0
  
Iterations4
Presolve Time0.00
Solution Time0.00


Simple Quadratic Program

This section contains Lua code for the analysis in the CASL version of this example, which contains details about the results.

Note: In order to run this code, the data that are described in the CASL version need to be accessible to the CAS server. One way to do this is to convert the example data to the comma-separated-value (CSV) file example.csv and then use the following code to load the CSV file into CAS:

s:loadtable{casLib="casuser", path="example.csv"}

For more information about coding in Lua, see Getting Started with SAS Viya for Lua and SAS Viya: System Programming Guide.

The following code solves the simple quadratic program stored in the data table example:

s:optimization_solveQp{
   data      = {name = "example"},
   primalOut = {name = "expout", replace=true},
   dualOut   = {name = "exdout", replace=true},
   objSense  = "min",
   logFreq   = 1}

Simple Quadratic Program

This section contains Python code for the analysis in the CASL version of this example, which contains details about the results.

Note: In order to run this code, the data that are described in the CASL version need to be accessible to the CAS server. One way to do this is to convert the example data to the comma-separated-value (CSV) file example.csv and then use the following code to load the CSV file into CAS:

s.upload_file('example.csv')

For more information about coding in Python, see Getting Started with SAS Viya for Python and SAS Viya: System Programming Guide.

The following code solves the simple quadratic program stored in the data table example:

s.optimization.solveQp(
    data      = {"name": "example"},
    primalOut = {"name": "expout", "replace":True},
    dualOut   = {"name": "exdout", "replace":True},
    objSense  = "min",
    logFreq   = 1)

Simple Quadratic Program

This section contains R code for the analysis in the CASL version of this example, which contains details about the results.

Note: In order to run this code, the data that are described in the CASL version need to be accessible to the CAS server. One way to do this is to convert the example data to the comma-separated-value (CSV) file example.csv and then use the following code to load the CSV file into CAS:

m <- s$upload("example.csv", casOut=list(name="example"))

For more information about coding in R, see Getting Started with SAS Viya for R and SAS Viya: System Programming Guide.

The following code solves the simple quadratic program stored in the data table example:

cas.optimization.solveQp(s,
   data      = "example",
   primalOut = list(name="expout", replace=TRUE),
   dualOut   = list(name="exdout", replace=TRUE),
   objSense  = "min",
   logFreq   = 1)
Last updated: April 22, 2022