java - Glpk java 输出

Question

如何在我的控制台或文件上打印 java-glpk 中所有已解决的问题？

类似这样的事情:(此输出来自 gusek)

------ ------------    ------------- ------------- -------------
     1 x[1]         *              1             0             1 
     2 x[2]         *              1             0             1 
     3 x[3]         *              0             0             1 
     4 x[4]         *              1             0             1 
     5 x[5]         *              1             0             1 
     6 x[6]         *              1             0             1 

我的java代码中的监听器(GlpkCallback.addListener(this)和GlpkTerminal.addListener(this))只返回这个:

GLPK Integer Optimizer, v4.55
3 rows, 15440 columns, 46320 non-zeros
15440 integer variables, all of which are binary
Preprocessing...
1 row, 15440 columns, 15440 non-zeros
15440 integer variables, all of which are binary
Scaling...
 A: min|aij| = 5.718e+001  max|aij| = 2.719e+005  ratio = 4.755e+003
GM: min|aij| = 1.000e+000  max|aij| = 1.000e+000  ratio = 1.000e+000
EQ: min|aij| = 1.000e+000  max|aij| = 1.000e+000  ratio = 1.000e+000
2N: min|aij| = 6.701e-001  max|aij| = 1.330e+000  ratio = 1.985e+000
Constructing initial basis...
Size of triangular part is 1
Solving LP relaxation...
GLPK Simplex Optimizer, v4.55
1 row, 15440 columns, 15440 non-zeros
*     0: obj =  1.584183802e+007  infeas = 0.000e+000 (0)
*   500: obj =  1.190189187e+007  infeas = 0.000e+000 (0)
*  1000: obj =  1.142465659e+007  infeas = 0.000e+000 (0)
*  1500: obj =  1.122247777e+007  infeas = 0.000e+000 (0)
*  1615: obj =  1.121676854e+007  infeas = 0.000e+000 (0)
OPTIMAL LP SOLUTION FOUND
Integer optimization begins...
+  1615: mip =     not found yet >=              -inf        (1; 0)
+  1615: >>>>>  1.121676854e+007 >=  1.121676854e+007   0.0% (1; 0)
Better solution found
+  1615: mip =  1.121676854e+007 >=     tree is empty   0.0% (0; 1)
INTEGER OPTIMAL SOLUTION FOUND

希望有人能帮忙。

这是我的java代码: https://jsfiddle.net/ilucasrds/enok1h5k/3/

Answer 1

在该示例中，您展示了两个不同的事物。一方面，监听器的输出包含求解器运行时提供的所有信息。另一方面，gusek 示例在求解器完成后提供了解决方案。

根据您想要的解决方案的信息类型，API 为您提供了四种不同的例程:

int glp_print_sol(glp_prob *P, const char *fname);
int glp_print_ipt(glp_prob *P, const char *fname);
int glp_print_mip(glp_prob *P, const char *fname);

用于 KKT 条件(单纯形/内部点)和整数可行性报告。

int glp_print_ranges(glp prob *P, int len, const int list[], int flags,
const char *fname);

写入敏感性分析报告，与 gusek 输出相当。通过附加的 len 和 list[] 属性，我们可以定义一组更具体的行/列来分析/输出。如果 len 为零，则分析每行/列。

为了进一步处理解决方案的代码，还可以使用

double glp_get_obj_coef(glp_prob *P, int j);

这将返回定义列的目标系数