diff --git a/AfterPreProcessing.txt b/AfterPreProcessing.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/bin/LagrangianRelaxation.linux.x86_64.gnu.opt.spx2 b/bin/LagrangianRelaxation.linux.x86_64.gnu.opt.spx2
index 5e83e6ad70999dad48a0743cdcd6053bfa4fbf60..3cc3f53e32b69aab6b8d4061ee8c57d4edbe8940 100755
Binary files a/bin/LagrangianRelaxation.linux.x86_64.gnu.opt.spx2 and b/bin/LagrangianRelaxation.linux.x86_64.gnu.opt.spx2 differ
diff --git a/dual.txt b/dual.txt
new file mode 100644
index 0000000000000000000000000000000000000000..9a989c716c2ec17c0854e245ef7d5f14751b0810
--- /dev/null
+++ b/dual.txt
@@ -0,0 +1,876 @@
+objective value: 2
+t_x_{1}_{0} 1 (obj:2)
+t_x_{2}_{0} 1 (obj:0)
+t_x_{3}_{0} 1 (obj:0)
+t_x_{4}_{0} 1 (obj:0)
+t_x_{5}_{0} 1 (obj:0)
+t_x_{6}_{0} 1 (obj:0)
+dualbound = 2.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 15
+t_x_{1}_{0} 1 (obj:5)
+t_x_{2}_{0} 1 (obj:3)
+t_x_{3}_{0} 1 (obj:4)
+t_x_{4}_{0} 1 (obj:4)
+t_x_{5}_{0} 1 (obj:4)
+t_x_{6}_{0} 1 (obj:2)
+dualbound = 22.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 28
+t_x_{1}_{0} 1 (obj:8)
+t_x_{2}_{0} 1 (obj:6)
+t_x_{3}_{0} 1 (obj:8)
+t_x_{4}_{0} 1 (obj:8)
+t_x_{5}_{0} 1 (obj:8)
+t_x_{6}_{0} 1 (obj:4)
+dualbound = 42.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 41
+t_x_{1}_{0} 1 (obj:11)
+t_x_{2}_{0} 1 (obj:9)
+t_x_{3}_{0} 1 (obj:12)
+t_x_{4}_{0} 1 (obj:12)
+t_x_{5}_{0} 1 (obj:12)
+t_x_{6}_{0} 1 (obj:6)
+dualbound = 62.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 54
+t_x_{1}_{0} 1 (obj:14)
+t_x_{2}_{0} 1 (obj:12)
+t_x_{3}_{0} 1 (obj:16)
+t_x_{4}_{0} 1 (obj:16)
+t_x_{5}_{0} 1 (obj:16)
+t_x_{6}_{0} 1 (obj:8)
+dualbound = 82.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 67
+t_x_{1}_{0} 1 (obj:17)
+t_x_{2}_{0} 1 (obj:15)
+t_x_{3}_{0} 1 (obj:20)
+t_x_{4}_{0} 1 (obj:20)
+t_x_{5}_{0} 1 (obj:20)
+t_x_{6}_{0} 1 (obj:10)
+dualbound = 102.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 80
+t_x_{1}_{0} 1 (obj:20)
+t_x_{2}_{0} 1 (obj:18)
+t_x_{3}_{0} 1 (obj:24)
+t_x_{4}_{0} 1 (obj:24)
+t_x_{5}_{0} 1 (obj:24)
+t_x_{6}_{0} 1 (obj:12)
+dualbound = 122.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 93
+t_x_{1}_{0} 1 (obj:23)
+t_x_{2}_{0} 1 (obj:21)
+t_x_{3}_{0} 1 (obj:28)
+t_x_{4}_{0} 1 (obj:28)
+t_x_{5}_{0} 1 (obj:28)
+t_x_{6}_{0} 1 (obj:14)
+dualbound = 142.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 100
+t_x_{1}_{0} 1 (obj:26)
+t_x_{2}_{0} 1 (obj:24)
+t_x_{3}_{15} 1 (obj:30)
+t_x_{4}_{15} 1 (obj:30)
+t_x_{5}_{15} 1 (obj:30)
+t_x_{6}_{0} 1 (obj:16)
+dualbound = 156.000000, lowerbound=46.000000, norm of subgrad 5.830952 objective value: 104
+t_x_{1}_{0} 1 (obj:27)
+t_x_{2}_{0} 1 (obj:25)
+t_x_{3}_{0} 1 (obj:32)
+t_x_{4}_{0} 1 (obj:32)
+t_x_{5}_{0} 1 (obj:32)
+t_x_{6}_{0} 1 (obj:16)
+dualbound = 164.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 102
+t_x_{1}_{0} 1 (obj:30)
+t_x_{2}_{0} 1 (obj:28)
+t_x_{3}_{15} 1 (obj:30)
+t_x_{4}_{15} 1 (obj:30)
+t_x_{5}_{15} 1 (obj:30)
+t_x_{6}_{0} 1 (obj:18)
+dualbound = 166.000000, lowerbound=46.000000, norm of subgrad 5.830952 objective value: 106
+t_x_{1}_{0} 1 (obj:31)
+t_x_{2}_{0} 1 (obj:29)
+t_x_{3}_{15} 1 (obj:32)
+t_x_{4}_{15} 1 (obj:32)
+t_x_{5}_{15} 1 (obj:32)
+t_x_{6}_{0} 1 (obj:18)
+dualbound = 174.000000, lowerbound=46.000000, norm of subgrad 5.830952 objective value: 110
+t_x_{1}_{0} 1 (obj:32)
+t_x_{2}_{0} 1 (obj:30)
+t_x_{3}_{15} 1 (obj:34)
+t_x_{4}_{15} 1 (obj:34)
+t_x_{5}_{15} 1 (obj:34)
+t_x_{6}_{0} 1 (obj:18)
+dualbound = 182.000000, lowerbound=46.000000, norm of subgrad 5.830952 objective value: 114
+t_x_{1}_{0} 1 (obj:33)
+t_x_{2}_{0} 1 (obj:31)
+t_x_{3}_{0} 1 (obj:36)
+t_x_{4}_{0} 1 (obj:36)
+t_x_{5}_{0} 1 (obj:36)
+t_x_{6}_{0} 1 (obj:18)
+dualbound = 190.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 106
+t_x_{1}_{15} 1 (obj:32)
+t_x_{2}_{15} 1 (obj:32)
+t_x_{3}_{15} 1 (obj:34)
+t_x_{4}_{15} 1 (obj:34)
+t_x_{5}_{15} 1 (obj:34)
+t_x_{6}_{0} 1 (obj:20)
+dualbound = 186.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 115
+t_x_{1}_{0} 1 (obj:34)
+t_x_{2}_{0} 1 (obj:32)
+t_x_{3}_{15} 1 (obj:38)
+t_x_{4}_{15} 1 (obj:37)
+t_x_{5}_{15} 1 (obj:37)
+t_x_{6}_{0} 1 (obj:20)
+dualbound = 198.000000, lowerbound=46.000000, norm of subgrad 5.830952 objective value: 118
+t_x_{1}_{0} 1 (obj:35)
+t_x_{2}_{0} 1 (obj:33)
+t_x_{3}_{0} 1 (obj:38)
+t_x_{4}_{0} 1 (obj:39)
+t_x_{5}_{0} 1 (obj:39)
+t_x_{6}_{0} 1 (obj:20)
+dualbound = 204.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 112
+t_x_{1}_{15} 1 (obj:34)
+t_x_{2}_{15} 1 (obj:34)
+t_x_{3}_{15} 1 (obj:38)
+t_x_{4}_{15} 1 (obj:37)
+t_x_{5}_{15} 1 (obj:37)
+t_x_{6}_{0} 1 (obj:22)
+dualbound = 202.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 119
+t_x_{1}_{0} 1 (obj:36)
+t_x_{2}_{0} 1 (obj:34)
+t_x_{3}_{0} 1 (obj:40)
+t_x_{4}_{15} 1 (obj:40)
+t_x_{5}_{15} 1 (obj:40)
+t_x_{6}_{0} 1 (obj:22)
+dualbound = 212.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 122
+t_x_{1}_{15} 1 (obj:36)
+t_x_{2}_{15} 1 (obj:36)
+t_x_{3}_{15} 1 (obj:42)
+t_x_{4}_{15} 1 (obj:41)
+t_x_{5}_{15} 1 (obj:41)
+t_x_{6}_{0} 1 (obj:22)
+dualbound = 218.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 117
+t_x_{1}_{0} 1 (obj:36)
+t_x_{2}_{0} 1 (obj:34)
+t_x_{3}_{0} 1 (obj:40)
+t_x_{4}_{0} 1 (obj:42)
+t_x_{5}_{0} 1 (obj:42)
+t_x_{6}_{0} 1 (obj:22)
+dualbound = 216.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 124
+t_x_{1}_{15} 1 (obj:37)
+t_x_{2}_{15} 1 (obj:37)
+t_x_{3}_{15} 1 (obj:44)
+t_x_{4}_{15} 1 (obj:42)
+t_x_{5}_{15} 1 (obj:42)
+t_x_{6}_{0} 1 (obj:24)
+dualbound = 226.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 123
+t_x_{1}_{0} 1 (obj:37)
+t_x_{2}_{0} 1 (obj:35)
+t_x_{3}_{0} 1 (obj:42)
+t_x_{4}_{0} 1 (obj:45)
+t_x_{5}_{0} 1 (obj:45)
+t_x_{6}_{0} 1 (obj:24)
+dualbound = 228.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 126
+t_x_{1}_{15} 1 (obj:38)
+t_x_{2}_{15} 1 (obj:38)
+t_x_{3}_{15} 1 (obj:46)
+t_x_{4}_{15} 1 (obj:43)
+t_x_{5}_{15} 1 (obj:43)
+t_x_{6}_{0} 1 (obj:26)
+dualbound = 234.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 125
+t_x_{1}_{0} 1 (obj:38)
+t_x_{2}_{0} 1 (obj:36)
+t_x_{3}_{0} 1 (obj:44)
+t_x_{4}_{15} 1 (obj:46)
+t_x_{5}_{15} 1 (obj:46)
+t_x_{6}_{0} 1 (obj:26)
+dualbound = 236.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 130
+t_x_{1}_{0} 1 (obj:40)
+t_x_{2}_{0} 1 (obj:38)
+t_x_{3}_{0} 1 (obj:46)
+t_x_{4}_{15} 1 (obj:47)
+t_x_{5}_{15} 1 (obj:47)
+t_x_{6}_{0} 1 (obj:26)
+dualbound = 244.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 132
+t_x_{1}_{15} 1 (obj:40)
+t_x_{2}_{0} 1 (obj:40)
+t_x_{3}_{0} 1 (obj:48)
+t_x_{4}_{15} 1 (obj:48)
+t_x_{5}_{15} 1 (obj:48)
+t_x_{6}_{0} 1 (obj:26)
+dualbound = 250.000000, lowerbound=45.000000, norm of subgrad 5.656854 objective value: 134
+t_x_{1}_{15} 1 (obj:41)
+t_x_{2}_{15} 1 (obj:40)
+t_x_{3}_{0} 1 (obj:49)
+t_x_{4}_{15} 1 (obj:49)
+t_x_{5}_{0} 1 (obj:50)
+t_x_{6}_{0} 1 (obj:26)
+dualbound = 255.000000, lowerbound=45.000000, norm of subgrad 5.830952 objective value: 135
+t_x_{1}_{0} 1 (obj:42)
+t_x_{2}_{0} 1 (obj:40)
+t_x_{3}_{0} 1 (obj:50)
+t_x_{4}_{15} 1 (obj:50)
+t_x_{5}_{15} 1 (obj:50)
+t_x_{6}_{0} 1 (obj:27)
+dualbound = 259.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 136
+t_x_{1}_{15} 1 (obj:41)
+t_x_{2}_{15} 1 (obj:41)
+t_x_{3}_{15} 1 (obj:52)
+t_x_{4}_{15} 1 (obj:51)
+t_x_{5}_{15} 1 (obj:51)
+t_x_{6}_{0} 1 (obj:27)
+dualbound = 263.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 133
+t_x_{1}_{0} 1 (obj:42)
+t_x_{2}_{0} 1 (obj:40)
+t_x_{3}_{0} 1 (obj:50)
+t_x_{4}_{0} 1 (obj:52)
+t_x_{5}_{0} 1 (obj:52)
+t_x_{6}_{0} 1 (obj:27)
+dualbound = 263.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 138
+t_x_{1}_{15} 1 (obj:42)
+t_x_{2}_{15} 1 (obj:42)
+t_x_{3}_{15} 1 (obj:54)
+t_x_{4}_{15} 1 (obj:52)
+t_x_{5}_{15} 1 (obj:52)
+t_x_{6}_{0} 1 (obj:29)
+dualbound = 271.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 139
+t_x_{1}_{0} 1 (obj:43)
+t_x_{2}_{0} 1 (obj:41)
+t_x_{3}_{0} 1 (obj:52)
+t_x_{4}_{0} 1 (obj:55)
+t_x_{5}_{0} 1 (obj:55)
+t_x_{6}_{0} 1 (obj:29)
+dualbound = 275.000000, lowerbound=1.000000, norm of subgrad 6.782330 objective value: 140
+t_x_{1}_{15} 1 (obj:43)
+t_x_{2}_{15} 1 (obj:43)
+t_x_{3}_{15} 1 (obj:56)
+t_x_{4}_{15} 1 (obj:53)
+t_x_{5}_{15} 1 (obj:53)
+t_x_{6}_{0} 1 (obj:31)
+dualbound = 279.000000, lowerbound=75.000000, norm of subgrad 6.480741 objective value: 141
+t_x_{1}_{0} 1 (obj:44)
+t_x_{2}_{0} 1 (obj:42)
+t_x_{3}_{0} 1 (obj:54)
+t_x_{4}_{15} 1 (obj:56)
+t_x_{5}_{15} 1 (obj:56)
+t_x_{6}_{0} 1 (obj:31)
+dualbound = 283.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 145
+t_x_{1}_{15} 1 (obj:45)
+t_x_{2}_{0} 1 (obj:44)
+t_x_{3}_{0} 1 (obj:56)
+t_x_{4}_{15} 1 (obj:57)
+t_x_{5}_{15} 1 (obj:57)
+t_x_{6}_{0} 1 (obj:31)
+dualbound = 290.000000, lowerbound=45.000000, norm of subgrad 5.656854 objective value: 148
+t_x_{1}_{15} 1 (obj:46)
+t_x_{2}_{15} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:57)
+t_x_{4}_{15} 1 (obj:58)
+t_x_{5}_{0} 1 (obj:59)
+t_x_{6}_{0} 1 (obj:31)
+dualbound = 296.000000, lowerbound=45.000000, norm of subgrad 5.830952 objective value: 147
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:44)
+t_x_{3}_{0} 1 (obj:58)
+t_x_{4}_{15} 1 (obj:59)
+t_x_{5}_{15} 1 (obj:59)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 298.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 150
+t_x_{1}_{15} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:46)
+t_x_{3}_{0} 1 (obj:60)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 304.000000, lowerbound=45.000000, norm of subgrad 5.656854 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952 objective value: 148
+t_x_{1}_{15} 1 (obj:47)
+t_x_{2}_{15} 1 (obj:46)
+t_x_{3}_{30} 1 (obj:60)
+t_x_{4}_{30} 1 (obj:60)
+t_x_{5}_{30} 1 (obj:60)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 305.000000, lowerbound=120.000000, norm of subgrad 5.830952 objective value: 146
+t_x_{1}_{0} 1 (obj:46)
+t_x_{2}_{0} 1 (obj:45)
+t_x_{3}_{0} 1 (obj:59)
+t_x_{4}_{15} 1 (obj:60)
+t_x_{5}_{15} 1 (obj:61)
+t_x_{6}_{0} 1 (obj:32)
+dualbound = 303.000000, lowerbound=31.000000, norm of subgrad 5.830952
\ No newline at end of file
diff --git a/lowerbounds.txt b/lowerbounds.txt
new file mode 100644
index 0000000000000000000000000000000000000000..78a0c6f4ca97f15ca8e5bf2587b5a8520e4c78b4
--- /dev/null
+++ b/lowerbounds.txt
@@ -0,0 +1,126 @@
+125
+1.000000
+1.000000
+1.000000
+1.000000
+1.000000
+1.000000
+1.000000
+1.000000
+46.000000
+1.000000
+46.000000
+46.000000
+46.000000
+1.000000
+75.000000
+46.000000
+1.000000
+75.000000
+31.000000
+75.000000
+1.000000
+75.000000
+1.000000
+75.000000
+31.000000
+31.000000
+45.000000
+45.000000
+31.000000
+75.000000
+1.000000
+75.000000
+1.000000
+75.000000
+31.000000
+45.000000
+45.000000
+31.000000
+45.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
+120.000000
+31.000000
diff --git a/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.d b/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.d
index f2767b4fafec7abb9ccd65f26e293b7fce7194c0..b5055b69b0805bc1bb473bac76f9ea1e2d6da137 100644
--- a/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.d
+++ b/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.d
@@ -86,4 +86,4 @@ obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.o: src/probdata_lagr.cpp \
../../src/scip/scip_solve.h ../../src/scip/scip_solvingstats.h \
../../src/scip/scip_table.h ../../src/scip/scip_timing.h \
../../src/scip/scip_tree.h ../../src/scip/scip_validation.h \
- ../../src/scip/scip_var.h src/vardata_lagr.h
+ ../../src/scip/scip_var.h src/vardata_lagr.h src/relax_lagr.h
diff --git a/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.o b/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.o
index bb713b419cb8f26e58ee6caebd6158c5da4e9b56..d8f8b75585f6cfe29fcadf2e41b9aa8a9dfcaebf 100644
Binary files a/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.o and b/obj/static/O.linux.x86_64.gnu.opt/probdata_lagr.o differ
diff --git a/obj/static/O.linux.x86_64.gnu.opt/relax_lagr.o b/obj/static/O.linux.x86_64.gnu.opt/relax_lagr.o
index de698773739367e637abb7408e08ff3ad7ec3c81..3efb4b2450cf7e55c1545e4712563848fc3259dd 100644
Binary files a/obj/static/O.linux.x86_64.gnu.opt/relax_lagr.o and b/obj/static/O.linux.x86_64.gnu.opt/relax_lagr.o differ
diff --git a/sol.txt b/sol.txt
new file mode 100644
index 0000000000000000000000000000000000000000..cec5f6155c2fca5b615bd14dda24cea8b5ca368a
--- /dev/null
+++ b/sol.txt
@@ -0,0 +1,251 @@
+number of solutions 1, first iteration bound=2.000000, objsol=2.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=15.000000, objsol=15.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=28.000000, objsol=28.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=41.000000, objsol=41.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=54.000000, objsol=54.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=67.000000, objsol=67.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=80.000000, objsol=80.000000
+lowerbound = 1.000000
+ number of solutions 1, first iteration bound=93.000000, objsol=93.000000
+lowerbound = 1.000000
+ number of solutions 2, first iteration bound=100.000000, objsol=100.000000
+lowerbound = 46.000000
+ number of solutions 2, first iteration bound=104.000000, objsol=104.000000
+lowerbound = 1.000000
+ number of solutions 2, first iteration bound=102.000000, objsol=102.000000
+lowerbound = 46.000000
+ number of solutions 2, first iteration bound=106.000000, objsol=106.000000
+lowerbound = 46.000000
+ number of solutions 2, first iteration bound=110.000000, objsol=110.000000
+lowerbound = 46.000000
+ number of solutions 2, first iteration bound=114.000000, objsol=114.000000
+lowerbound = 1.000000
+ number of solutions 3, first iteration bound=106.000000, objsol=106.000000
+lowerbound = 75.000000
+ number of solutions 3, first iteration bound=115.000000, objsol=115.000000
+lowerbound = 46.000000
+ number of solutions 3, first iteration bound=118.000000, objsol=118.000000
+lowerbound = 1.000000
+ number of solutions 3, first iteration bound=112.000000, objsol=112.000000
+lowerbound = 75.000000
+ number of solutions 4, first iteration bound=119.000000, objsol=119.000000
+lowerbound = 31.000000
+ number of solutions 4, first iteration bound=122.000000, objsol=122.000000
+lowerbound = 75.000000
+ number of solutions 4, first iteration bound=117.000000, objsol=117.000000
+lowerbound = 1.000000
+ number of solutions 4, first iteration bound=124.000000, objsol=124.000000
+lowerbound = 75.000000
+ number of solutions 4, first iteration bound=123.000000, objsol=123.000000
+lowerbound = 1.000000
+ number of solutions 4, first iteration bound=126.000000, objsol=126.000000
+lowerbound = 75.000000
+ number of solutions 4, first iteration bound=125.000000, objsol=125.000000
+lowerbound = 31.000000
+ number of solutions 4, first iteration bound=130.000000, objsol=130.000000
+lowerbound = 31.000000
+ number of solutions 5, first iteration bound=132.000000, objsol=132.000000
+lowerbound = 45.000000
+ number of solutions 6, first iteration bound=134.000000, objsol=134.000000
+lowerbound = 45.000000
+ number of solutions 6, first iteration bound=135.000000, objsol=135.000000
+lowerbound = 31.000000
+ number of solutions 6, first iteration bound=136.000000, objsol=136.000000
+lowerbound = 75.000000
+ number of solutions 6, first iteration bound=133.000000, objsol=133.000000
+lowerbound = 1.000000
+ number of solutions 6, first iteration bound=138.000000, objsol=138.000000
+lowerbound = 75.000000
+ number of solutions 6, first iteration bound=139.000000, objsol=139.000000
+lowerbound = 1.000000
+ number of solutions 6, first iteration bound=140.000000, objsol=140.000000
+lowerbound = 75.000000
+ number of solutions 6, first iteration bound=141.000000, objsol=141.000000
+lowerbound = 31.000000
+ number of solutions 6, first iteration bound=145.000000, objsol=145.000000
+lowerbound = 45.000000
+ number of solutions 6, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 45.000000
+ number of solutions 6, first iteration bound=147.000000, objsol=147.000000
+lowerbound = 31.000000
+ number of solutions 6, first iteration bound=150.000000, objsol=150.000000
+lowerbound = 45.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+ number of solutions 7, first iteration bound=148.000000, objsol=148.000000
+lowerbound = 120.000000
+ number of solutions 7, first iteration bound=146.000000, objsol=146.000000
+lowerbound = 31.000000
+
\ No newline at end of file
diff --git a/src/probdata_lagr.cpp b/src/probdata_lagr.cpp
index cf583765177aeb2f75e35a22e57e68414821c919..32eab2dd2583f967608cff79af88e0331b523064 100644
--- a/src/probdata_lagr.cpp
+++ b/src/probdata_lagr.cpp
@@ -6,6 +6,7 @@
#include <assert.h>
#include <stdlib.h>
#include <string.h>
+#include "relax_lagr.h"
/*Using ProbData as a memory location for the constraints*/
struct SCIP_ProbData
{
@@ -14,11 +15,14 @@ struct SCIP_ProbData
int* varids; /**<Ids for the variables found in the slot constraint*/
int nSlotConss; // number of slot constraints.
int nStartConss;
- SCIP_Real* upperbound;
+ SCIP_Real* upperbound;
int nbadconss;
int ngoodconss;
int* badconss;
int* goodconss;
+ int allnconsvars;
+ int* listnconsvars;
+ int* listconsvarids;
};
@@ -28,6 +32,18 @@ SCIP_ProbData* probdata
return probdata->varids;
}
+int* SCIPlistnconsvars(
+SCIP_ProbData* probdata
+){
+ return probdata->listnconsvars;
+ }
+
+int* SCIPlistconsvarids(
+SCIP_ProbData* probdata
+){
+ return probdata->listconsvarids;
+ }
+
SCIP_CONS** SCIPgetSlotConss(
SCIP_ProbData* probdata
){
@@ -40,6 +56,12 @@ SCIP_ProbData* probdata
return probdata->nSlotConss;
}
+int SCIPgetallnconsvars(
+SCIP_ProbData* probdata
+){
+ return probdata->allnconsvars;
+ }
+
SCIP_RETCODE GetNGoodandNbad(
SCIP* scip,
int* nbad,
@@ -138,43 +160,176 @@ SCIP_RETCODE probdataFree(
/*We will create the probdata to have the slot constraints and the variables they hold withn them*/
SCIP_RETCODE SCIPcreateprobdata
-( SCIP* scip,
- SCIP_ProbData** probdata
+( SCIP* relaxscip,
+ SCIP_ProbData** probdata,
+ SCIP_CONS** conss,
+ SCIP_VAR** vars,
+ SCIP_VAR*** varbuffers,
+ int** badconss
)
{
- SCIP_CONS** conss = SCIPgetConss(scip);
- int nconss = SCIPgetNConss(scip);
+ int nconss = SCIPgetNConss(relaxscip);
+ int nvars = SCIPgetNVars(relaxscip);
int nSlotConss = 0;
-
+ int id =0;
SCIP_Bool success;
for (int t=0; t<nconss; t++) /* (3) */
{
-
SCIP_CONS* cons = conss[t];
- if(SCIPconsGetLhs(scip,cons,&success)==-SCIPinfinity(scip)) /*<We get the slot constraints based on the inquality*/
+ if(SCIPconsGetLhs(relaxscip,cons,&success)==-SCIPinfinity(relaxscip)) /*<We get the slot constraints based on the inquality*/
{
- // printf("%s)",SCIPconsGetName(cons));
++nSlotConss;
}
- //* To make sure that there is no misarrangmenets, we make sure that the slot and start constraints are clearly separated*//
- // else if(SCIPconsGetLhs(scip,cons,&success)==SCIPinfinity(scip))
- // {
- // for (int i = t; i < nconss; i++)
- // {
- // SCIP_CONS* nextcons = conss[i+1];
- // if (SCIPconsGetLhs(scip,nextcons,&success)==-SCIPinfinity(scip))
- // {
- // printf("oops");
- // SCIPerrorMessage("\n********\tLagrangian Relaxation CAN'T BE USED!\t********* \n The problem has mixed it's inequality and equality(bad and good) constraints. Please rearrange with first bad then good\n");
- // SCIPABORT();
- // break;
- // }
-
- // }
- // }
}
- SCIP_CALL(SCIPallocMemory(scip,probdata));
+ printf("%d nslot conss",nSlotConss);
+
+ int allnconsvars=0;
+ int nconsvars=0;
+ int counter = 0;
+ int maxnconsvars = 0;
+
+
+ SCIP_CALL(SCIPallocBufferArray(relaxscip, badconss, nSlotConss)); //the badconss array will contain the row number of the bad conss
+
+ /*
+ we first get the row numbers of the bad conss, and save to the array: badconss.
+ */
+ for(int r = 0; r<nconss; ++r)
+ {
+ SCIP_CONS* cons = conss[r];
+
+ if(SCIPconsGetLhs(relaxscip,cons,&success)==-SCIPinfinity(relaxscip)) /*<We get the slot constraints based on the inquality*/
+ {
+ (*badconss)[counter]=r;
+ SCIP_CALL(SCIPgetConsNVars(relaxscip,cons,&nconsvars,&success));
+ counter++;
+ if(maxnconsvars<nconsvars){maxnconsvars=nconsvars;}
+ allnconsvars+=nconsvars;
+ }
+ }
+
+ /*
+ Our first objective is to create an array, containing the non-zero variables
+ found in each bad conss. But instead of listing them in multiple rows, we list them just in one.
+ for example: slot1 have vars[0] and vars[10] non-zero and slot2 has vars[1] and vars[11].
+ listconsvarids = {0,10,1,11, ...}
+ */
+ int* listconsvarids; //examples {0,10,1,11,2,12,...} we use the
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&listconsvarids,allnconsvars));
+
+ /*
+ we save the number of non-zero variables in bad constraint has.
+ */
+ int* listnconsvars; //example {2,2,2,3,3,3}
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&listnconsvars,nSlotConss));
+
+ /*to not allocate buffer array over and over again, we create one, with
+ size = maxnconsvars. and then we intialize it with the first few variables from vars.
+ */
+ SCIP_CALL(SCIPallocBufferArray(relaxscip, varbuffers, maxnconsvars));
+ for (int v = 0; v < maxnconsvars; v++)
+ {
+ (*varbuffers)[v] =vars[v];
+ }
+
+
+ counter=0;
+ for (int r = 0; r < nSlotConss; ++r)
+ {
+ nconsvars=0;
+ id = (*badconss)[r];
+ SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+
+ // printf("%s \t",SCIPconsGetName(cons));
+ SCIP_CALL(SCIPgetConsNVars(relaxscip, cons, &nconsvars, &success));
+ SCIP_CALL(SCIPgetConsVars(relaxscip, cons, (*varbuffers), nconsvars, &success));
+ if (!success){
+ abort(); }
+ listnconsvars[r] = nconsvars;
+ for (int j = 0; j < nconsvars; ++j) /* (8) */
+ {
+ SCIP_VAR* varx = (*varbuffers)[j];
+ int varbufindex = SCIPvarGetIndex(varx);
+ assert(varbufindex != NULL);
+ listconsvarids[counter]=varbufindex;
+ counter++;
+ }
+ }
+
+ // counter=0;
+ // for(int r=0; r<nSlotConss;++r)
+ // {
+ // for(int s=counter;s<(counter+listnconsvars[r]);++s)
+ // {
+ // printf("%s->",SCIPvarGetName(vars[listconsvarids[s]]));
+ // }
+ // printf("\n");
+ // counter += listnconsvars[r];
+ // // printf("%d\n->",counter);
+ // }
+ int* listvarswconsid;
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&listvarswconsid,allnconsvars));
+
+ //* use the following to create an alternative way of varids*//
+ // int* listvarncons;
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip,&listvarncons,nvars));
+
+ // int count;
+ // for(int v=0; v<nvars;++v)
+ // {
+ // int varid = SCIPvarGetIndex(vars[v]);
+ // counter=0;
+ // count =0;
+
+ // // printf("%s->",SCIPvarGetName(vars[v]));
+ // for(int r=0; r<nSlotConss;++r)
+ // {
+
+ // id = (*badconss)[r];
+ // SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // // printf("%s \t",SCIPconsGetName(cons));
+ // for(int s=counter;s<(counter+listnconsvars[r]);++s)
+ // {
+ // // printf("(%d)",s);
+ // if(varid==listconsvarids[s])
+ // {
+ // listvarswconsid[s]=id;
+ // count++;
+ // // printf("%s->",SCIPconsGetName(conss[listvarswconsid[s]]));
+ // }
+ // }
+ // counter += listnconsvars[r];
+ // // printf("%d\n->",counter);
+ // }
+ // listvarncons[v]=count;
+ // // printf("\n");
+ // }
+
+
+
+ // counter=0;
+ // for(int v=0; v<nvars;++v)
+ // {
+ // SCIP_VAR* var = vars[v];
+ // printf("%s,%d",SCIPvarGetName(var),listvarncons[v]);
+ // //next step is a creation of an array that will tell in how many bad constraints a variable is found in.
+ // for(int s=counter;s<(counter+listvarncons[v]);++s)
+ // {
+ // // printf("%s->",SCIPconsGetName(conss[listconsvarids[s]]));
+ // printf("%s,",SCIPconsGetName(conss[listconsvarids[s]]));
+ // }
+ // printf("\n");
+ // counter += listvarncons[v];
+ // // // printf("%d\n->",counter);
+ // }
+
+
+
+ SCIP_CALL(SCIPallocMemory(relaxscip,probdata));
(*probdata)->nSlotConss = nSlotConss;
+ (*probdata)->allnconsvars= allnconsvars; //sum of all nconsvars;
+ (*probdata)->listnconsvars = listnconsvars;
+ (*probdata)->listconsvarids = listconsvarids;
return SCIP_OKAY;
}
@@ -269,7 +424,7 @@ SCIP_Real SCIPgetSubgradients(
SCIP_Bool success;
int nconsvars;
SCIP_VAR** varbuffer;
- FILE* filos;
+
// SCIP_SOL* bestsol = SCIPgetBestSol(relaxscip);
// SCIP_CALL(SCIPprintSol(relaxscip,bestsol,filos,FALSE));
@@ -334,8 +489,6 @@ SCIP_RETCODE SCIPsolveiteration(SCIP* scip,int nSlotConss,SCIP_Real** subgradien
int nvars = SCIPgetNVars(scip);
// SCIP_Real C= 2;
- FILE* file;
-
SCIP_CALL( SCIPcreate(&lagrscip));
SCIP_Bool valid=FALSE;
// SCIP_CALL( SCIPhashmapCreate(&varmap, SCIPblkmem(lagrscip), nvars) );
@@ -357,7 +510,6 @@ SCIP_RETCODE SCIPsolveiteration(SCIP* scip,int nSlotConss,SCIP_Real** subgradien
/*get the best solution*/
SCIP_SOL* bestsol ;
- SCIP_SOL** sols;
SCIP_Real* solvals;
SCIP_CALL(SCIPallocBufferArray(lagrscip,&solvals,nvars+1));
@@ -375,7 +527,6 @@ SCIP_RETCODE SCIPsolveiteration(SCIP* scip,int nSlotConss,SCIP_Real** subgradien
//prinf("\n");
for (int v = 0; v<nvars-1; ++v)
{
- SCIP_VAR* var = SCIPgetVars(lagrscip)[v];
if( solvals[v] == 1 && solvals[v+1] == 1)
{
solvals[v] = 0;
@@ -387,7 +538,6 @@ SCIP_RETCODE SCIPsolveiteration(SCIP* scip,int nSlotConss,SCIP_Real** subgradien
}
/*to get a better upperbound*/
- SCIP_Real solobj;
// SCIPgetFeasUpperbound(scip,&secondsolvals,&upperbound, &solobj, niter);
/*if the obj value in this iteration is bette(in our case, higher), than the previous one, save the solution array and also it's obj value to the bestsolvals*/
@@ -490,3 +640,76 @@ SCIP_RETCODE SCIPgetFeasUpperbound(SCIP* scip, SCIP_Real** solvals, SCIP_Real**
return SCIP_OKAY;
}
+SCIP_RETCODE scipgetsolutions(SCIP* scip, SCIP_VAR** vars, SCIP_Real** solvals, SCIP_Real* relaxval, SCIP_Real* dualmultipliers, SCIP_Real sumofduals,SCIP_SOL** bestsol)
+{
+ double sum;
+ FILE* varobj;
+ varobj=fopen("varobj.txt","wr");
+ FILE* problemstate;
+ problemstate = fopen("problemstate.txt","w+");
+ FILE* solutions;
+ solutions = fopen("solutions2.txt","w+");
+
+ SCIP_VARDATA* vardata;
+
+ int nvars = SCIPgetNVars(scip);
+ for(int v=0;v<nvars;v++)
+ {
+ SCIP_VAR* var = vars[v];
+ sum =SCIPvarGetObj(var);
+
+ vardata=SCIPvarGetData(var);
+ int* varids = SCIPvardataGetvarids(vardata);
+ int NVarInBadConss = SCIPvardataGetNVarInBadConss(vardata);
+
+ for(int t=0;t<NVarInBadConss;t++)
+ {
+ sum += dualmultipliers[varids[t]];
+ fprintf(varobj,"{%d, %f, %f\t",varids[t], dualmultipliers[varids[t]],sum);
+ }
+
+
+ // findmin += sum*solvals[v];
+ // fprintf(varobj, "solval %f, coefficient %f, sum %f", solvals[v],sum, findmin);
+ fprintf(varobj,"}\n\n");
+ SCIP_CALL(SCIPaddVarObj(scip,var,sum));
+ // add = weights[v]+sum;
+
+ }
+
+ // SCIPinfoMessage(scip, TimeCollector, "\n finished changing the variable's weight after (sec) : %f\n", SCIPgetClockTime(scip, varslottime));
+
+ SCIP_CALL(SCIPaddOrigObjoffset(scip,-1*sumofduals));
+ SCIP_CALL(SCIPprintOrigProblem(scip, problemstate, "lp", FALSE));
+ SCIPsetMessagehdlrQuiet(scip, TRUE);
+ // fclose(AfterPreProcessing);
+
+ SCIP_CALL( SCIPtransformProb(scip) );
+ SCIP_CALL( SCIPsolve(scip) );
+ *relaxval = SCIPgetPrimalbound(scip);
+ SCIPdebugMessage("relaxation bound = %e status = %d\n", *relaxval, SCIPgetStatus(scip));
+ /*get the best solution*/
+ *bestsol = SCIPgetBestSol(scip) ;
+ SCIP_CALL(SCIPallocBufferArray(scip,solvals,nvars+1));
+
+ /*text output*/
+ fprintf(solutions,"first iteration \t bound=%f, \t objsol=%f \n",SCIPgetPrimalbound(scip),*relaxval);
+ SCIP_CALL(SCIPprintBestSol(scip,solutions,FALSE));
+
+ /*store the solution in solvals so we can later export it to subgradient function*/
+ SCIPgetSolVals(scip,*bestsol,nvars,vars,*solvals);
+ fclose(varobj);
+ fclose(solutions);
+ fclose(problemstate);
+}
+
+SCIP_Real getnorm(SCIP_Real* array, int sizeofarray, SCIP_Real stepsize)
+{
+ SCIP_Real norm;
+ for(int r=0; r<sizeofarray;++r)
+ {
+ norm += array[r]*array[r];
+ }
+ norm=sqrt(norm);
+ return norm;
+}
\ No newline at end of file
diff --git a/src/probdata_lagr.h b/src/probdata_lagr.h
index 11deabefff3a7d25f1cd397d77b4efd14e04f51f..de1c397f937960df4597e4e641191138c21fb5b0 100644
--- a/src/probdata_lagr.h
+++ b/src/probdata_lagr.h
@@ -16,10 +16,17 @@
/*classifying and storing the slot and start constraints in PROBDATA
create probdata*/
SCIP_RETCODE SCIPcreateprobdata(
- SCIP* scip,
- SCIP_ProbData** probdata
- );
+ SCIP* scip,
+ SCIP_ProbData** probdata,
+ SCIP_CONS** conss,
+ SCIP_VAR** vars,
+ SCIP_VAR*** varbuffers,
+ int** badconss
+);
+int SCIPgetallnconsvars(
+SCIP_ProbData* probdata
+ );
SCIP_RETCODE GetNGoodandNbad(
SCIP* scip,
int* nbad,
@@ -32,12 +39,30 @@ SCIP_RETCODE GetNGoodandNbad(
int* SCIPconsGetvarids(
SCIP_ProbData* probdata
);
+int SCIPgetallmaxnconsvars(
+SCIP_ProbData* probdata
+);
+int* SCIPlistnconsvars(
+SCIP_ProbData* probdata
+);
+
+SCIP_Real getnorm(SCIP_Real* array, int sizeofarray, SCIP_Real stepsize);
+
+int* SCIPlistconsvarids(
+SCIP_ProbData* probdata
+);
SCIP_CONS** SCIPgetSlotConss(
SCIP_ProbData* probdata
);
-
+SCIP_RETCODE scipgetsolutions(
+ SCIP* scip, SCIP_VAR** vars,
+ SCIP_Real** solvals,
+ SCIP_Real* relaxval,
+ SCIP_Real* dualmultipliers,
+ SCIP_Real sumofduals,
+ SCIP_SOL** bestsol);
int SCIPgetNSlotConss(
SCIP_ProbData* probdata
diff --git a/src/relax_lagr.cpp b/src/relax_lagr.cpp
index f7e1679b0fbbd4632329f993b0ffe0db023ae647..212e44c1c596f0358cdc9d7715d49888c6a75591 100644
--- a/src/relax_lagr.cpp
+++ b/src/relax_lagr.cpp
@@ -61,6 +61,7 @@ struct SCIP_RelaxData
{
SCIP_SOL* sol; /**current solution(working solution)*/
SCIP_VARDATA* vardata;
+ SCIP_CONSDATA* consdata;
SCIP_Real* bestsolvals;
SCIP_Real* feasiblesol;
SCIP_Real* upperbound;
@@ -72,10 +73,10 @@ struct SCIP_VarData
SCIP_CONS** VarConss;
int nVarConss;
SCIP_CONS** VarSlotConss; /**<contains all slot constraints containing the variable */
- int nVarSlotConss; /**<number of slot constraints the variable is occuring in*/
+ int NVarInBadConss; /**<number of slot constraints the variable is occuring in*/
SCIP_Real varquotient;
int* consids;
- int* varid;
+ int* varids;
};
@@ -94,236 +95,400 @@ SCIP_DECL_RELAXFREE(relaxFreelagr)
/** initialization method of relaxator (called after problem was transformed) */
-static
-SCIP_DECL_RELAXINIT(relaxInitlagr)
-{ /*lint --e{715}*/
- printf("hellow\n");
+int SCIPvardataGetNVarInBadConss(
+ SCIP_VARDATA* vardata /**< variable data */
+)
+ {
+ return vardata->NVarInBadConss;
+ }
- using std::cout;
- SCIP* relaxscip; //a copy of the scip we can work on
- SCIP_HASHMAP* varmap; //to map the variables from scip to relaxscip
- SCIP_HASHMAP* consmap; //to map the constraints from scip to relaxscip
+int* SCIPvardataGetvarids(
+ SCIP_VARDATA* vardata /**< variable data */
+)
+ {
+ return vardata->varids;
+ }
- SCIP_VARDATA* vardata; //contains variable data, including in which constraints it's found
- SCIP_PROBDATA* probdata; //contains constraint data, identify slot vs start conss, and also which variables it contains
- SCIP_RELAXDATA* relaxdata; // not yet used, but can be part of the struct SCIP_RelaxData
- SCIP_CLOCK* varslottime; //to help us record the time
- FILE* file;
+static
+SCIP_DECL_RELAXINIT(relaxInitlagr)
+{ /*lint --e{715}*/
- SCIP_Bool success;
- SCIP_Bool valid;
-
- SCIP_SOL** sols;
-
- relaxdata = SCIPrelaxGetData(relax);
- SCIP_CALL(SCIPallocBlockMemory(scip , &relaxdata));
- assert(relaxdata != NULL);
+ SCIP* relaxscip;
+ SCIP_HASHMAP* varmap;
+ SCIP_HASHMAP* consmap;
+ SCIP_CONS** conss;
+ SCIP_PROBDATA* probdata;
+ SCIP_VARDATA* vardata;
+ SCIP_Real relaxval;
+ SCIP_Bool valid;
+ int nconss;
+ int i;
+ int counter;
+ int id;
+ // *lowerbound = -SCIPinfinity(scip);
+ // *result = SCIP_DIDNOTRUN;
+ /* we can only run if none of the present constraints expect their variables to be binary or integer during transformation */
+ conss = SCIPgetConss(scip);
+ nconss = SCIPgetNConss(scip);
- /*************************************************************************************************/
- /*we start with creating our relaxed scip instance, where the variable and constraints are mapped*/
- /*************************************************************************************************/
- SCIP_CALL( SCIPcreate(&relaxscip));
- valid=FALSE;
+ /* create the variable mapping hash map */
+ SCIP_CALL( SCIPcreate(&relaxscip) );
SCIP_CALL( SCIPhashmapCreate(&varmap, SCIPblkmem(relaxscip), SCIPgetNVars(scip)) );
- SCIP_CALL(SCIPcopy(scip,relaxscip,varmap,consmap,"relaxscip",FALSE,FALSE,FALSE,FALSE,&valid));
+ valid = FALSE;
+ SCIP_CALL( SCIPcopy(scip, relaxscip, varmap, consmap, "relaxscip", FALSE, FALSE, FALSE, FALSE, &valid) );
- SCIP_VAR** origvars = SCIPgetVars(scip);
- SCIP_VAR** vars = SCIPgetVars(relaxscip);
- SCIP_CONS** conss = SCIPgetConss(relaxscip);
- int nconss = SCIPgetNConss(relaxscip);
- int nvars = SCIPgetNVars(relaxscip);
- SCIP_Real upperbound = 0;
- // printf("**************%f***********",SCIPgetNCountedFeasSubtrees(relaxscip));
-
- SCIP_CALL( SCIPcreateClock(relaxscip, &varslottime) ); //* start time counting*
- SCIP_CALL(SCIPstartClock(relaxscip,varslottime));
-
/**************************************************************************************************************/
/*First, */
//*the probdata: where we get to identify the bad constraint we want to formulate(in our case, the slot conss) */
- //*the vardata: where we will get save the slot constraints on which the variable exists, */
- /*--along with the quotient of the obj to the nVarSlotConss */
/***************************************************************************************************************/
-
- SCIPcreateprobdata(relaxscip,&probdata); //*will be used to identify the #of slot constraints
- printf("%d", SCIPgetNConss(relaxscip));
- int nSlotConss = SCIPgetNSlotConss(probdata);
- printf("%d", SCIPgetNSlotConss(probdata));
+ int nvars = SCIPgetNVars(relaxscip);
+ SCIP_VAR** vars = SCIPgetVars(relaxscip);
+ SCIP_VAR** varbuffers;
+ int* badconss;
+
+ SCIPcreateprobdata(relaxscip,&probdata,SCIPgetConss(relaxscip),vars,&varbuffers,&badconss); /*will be used to identify the # of slot(bad) constraints*/
+ int nSlotConss = SCIPgetNSlotConss(probdata); //number of bad(slot) constraint
+ int allnconsvars = SCIPgetallnconsvars(probdata); //sum of all nconsvars, used for creating later on an array to collect the list of varids in each row
+ int* listnconsvars = SCIPlistnconsvars(probdata);
+ int* listconsvarids = SCIPlistconsvarids(probdata);
+
+ /* we then create the vardata function for each variable, to see at which constraint the variable is found*/
+ FILE* TimeCollector;
+ TimeCollector = fopen("time.txt","w");
+ SCIP_CLOCK* varslottime; //to help us record the time
+ SCIP_CALL( SCIPcreateClock(relaxscip, &varslottime) ); //* start time counting*
+ SCIP_CALL(SCIPstartClock(relaxscip,varslottime));
+
+ // int nconsvars=0;
+ int* consids;
- SCIP_Real* origobj;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&origobj,nvars));
+ SCIP_Real* weights;
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&weights,nvars));
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&consids,nSlotConss));
for (int v = 0; v < nvars; v++)
{
SCIP_VAR* var = vars[v];
- origobj[v]= SCIPvarGetObj(var);
+ weights[v]=SCIPvarGetObj(var);
+ }
+
+ for (int v = 0; v < nvars; v++)
+ {
+ int* varids;
+ int NVarInBadConss=0;
+ int nconsvars = 0;
+ SCIP_VAR* var = vars[v];
+
+ int varindex = SCIPvarGetIndex(var); /* (2) */
+ assert(varindex!= NULL);
- vardata=NULL;
- SCIPvardataCreateLagrangian(relaxscip,vardata,&var,nSlotConss,v); //*we create the vardata, containing the ids of the constraints containing the variable
- vardata=SCIPvarGetData(var);
- if(upperbound<=SCIPvarGetObj(var))
+ // printf("%s****%d\n",SCIPvarGetName(var),varindex);
+ for (int r = 0; r < nSlotConss; ++r)
{
- upperbound = SCIPvarGetObj(var);
+ id = badconss[r];
+ SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // printf("%s \t",SCIPconsGetName(cons));
+ SCIP_CALL(SCIPgetConsNVars(relaxscip, cons, &nconsvars, &valid));
+ SCIP_CALL(SCIPgetConsVars(relaxscip, cons, varbuffers, nconsvars, &valid));
+ if (!valid){
+ abort(); }
+
+ for (int j = 0; j < nconsvars; ++j) /* (8) */
+ {
+ SCIP_VAR* varx = varbuffers[j];
+ int varbufindex = SCIPvarGetIndex(varx);
+ assert(varbufindex != NULL);
+ // printf("%s\t \t%d",SCIPvarGetName(varx),varbufindex);
+
+
+ /** if var[i] is in cons[c], write conspointer in VarConss and increase nVarConsscounter */
+ if (varindex == varbufindex) /* (9) */
+ {
+
+ // VarSlotConss[NVarInBadConss] = cons;
+ consids[NVarInBadConss]=id;
+ NVarInBadConss++;
+ // printf(" %s \t,",SCIPconsGetName(cons));
+ }
+ }
+ }
+
+ SCIP_CALL(SCIPallocBufferArray(relaxscip, &varids, NVarInBadConss));
+ for(int t=0;t<NVarInBadConss;t++)
+ {
+ varids[t]=consids[t];
+ // printf("%d \t",varids[t]);
}
- }
- upperbound = upperbound * (nconss - nSlotConss);
- printf("%f",upperbound);
- // SCIP_CALL(SCIPstopClock(relaxscip,varslottime));
- // SCIPinfoMessage(relaxscip, NULL, "Solving Time (sec) : %6f\n", SCIPgetClockTime(relaxscip, varslottime));
+ // vardata=SCIPvarGetData(var);
+ SCIP_CALL(SCIPallocBlockMemory(scip , &vardata));
+ SCIP_CALL(SCIPduplicateBlockMemoryArray(scip, &(vardata->varids), varids, NVarInBadConss));
+ vardata->NVarInBadConss = NVarInBadConss; /**copy nVarConss to VarData */
+ vardata->varids = varids;
+ // /**set the variable data to the variable*/
+ SCIPvarSetData(var,vardata);
+ }
- SCIP_SOL* bestsol ;
- SCIP_Real* dualmultipliers;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&dualmultipliers,nSlotConss));
-
- SCIP_Real sumofduals = 0;
- SCIP_Real stepsize = 0.0000;
- SCIP_Real sqsum = 0;
- SCIP_Real sqrtsum = 0;
- SCIP_Real C = 2;
+ // SCIP_CALL(SCIPstopClock(relaxscip,varslottime));
+
+ FILE* AfterPreProcessing;
+ AfterPreProcessing = fopen("AfterPreProcessing.txt","w+");
- SCIP_Real* subgradients;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&subgradients,nSlotConss));
+ // SCIP_CALL(SCIPprintOrigProblem(relaxscip, AfterPreProcessing, "lp", FALSE));
+
+ SCIPinfoMessage(relaxscip, TimeCollector, "\n row and column identified in (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ for(int r=0;r<nSlotConss;r++)
+ {
+ id = badconss[r];
+ SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ SCIP_CALL(SCIPdelCons(relaxscip,cons));
+ }
/******************************************************************************************************************/
- /*Next, we will do the initial iteration of finding the dual mulpliers of each slot conss, and their sum(dualsum) */
+ /*Next, we will do the initial iteration of finding the dual mulpliers of each slot conss, and their sum(dualsum) */
/* In the end, we will subtract this sum from the objective of the function. */
/* It's initial, because while we would search for more dual multipliers to solve the Lagrangian relaxation */
/******************************************************************************************************************/
-
+ SCIP_Real* dualmultipliers;
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&dualmultipliers,nSlotConss));
+
+ SCIP_Real* subgradients;
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&subgradients,nSlotConss));
+ //initialize subgradients;
+ SCIP_Real stepsize = 1.00000;
+ SCIP_Real sumofduals=0;
for ( int r = 0; r < nSlotConss; ++r)
{
- // printf("*******************%f\n",subgradients[r]);
- SCIP_CONS* cons = conss[r];
-
- dualmultipliers[r]=SCIPconsGetMultiplier(relaxscip,&cons, 0,0,0,true,0); //returns the minumum
+ // id = badconss[r];
+ // SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ //if k=1 iteration//
dualmultipliers[r] = 0;
sumofduals+=dualmultipliers[r]; //adds the negative of the minimum in each iteration
- // printf("(%s,%f) \n",SCIPconsGetName(cons), dualmultipliers[r]);
-
+
}
-
- int niters = 2250;
- int stoppingiter=0;
-
- /**solvals is used for storing solutions in an iteration. where as bestsolvals will store the solution with the best objective value*/
- SCIP_Real* solvals;
- SCIP_Real* bestsolvals;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&solvals,nvars+1));
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&bestsolvals,nvars+1+niters));
- bestsolvals[nvars] = -10000000000000000;
- // printf("%f",bestsolvals[nvars+1]);
+
+ /*******************************************************************************************************/
+ /* The reformulation of the problem can be written as follows */
+ //*>>>>>>>>>>>>>>>>>> min sum { (w[i]+sum{dual[j]})}x[i]-sum{dual[r]} <<<<<<<<<<<< */
+ /*where i is nvars, j is NVarInBadConss, and r is nSlotConss for our case *******************************/
+ /****************************************************************************************************************/
+ /* The following function will add the following to the obj(weight) of the variable, */
+ //* the obj(weight) of var + the sum of the dualmultipliers of bad constraints which contains this variable */
+ /****************************************************************************************************************/
+
+
+ FILE* solutions;
+ solutions = fopen("sol.txt","wr");
+ FILE* dual;
+ dual= fopen("dual.txt","wr");
+ FILE* variableinfo;
+ variableinfo = fopen("var.txt","wr");
+ FILE* subgrad;
+ subgrad = fopen("subgrads.txt","wr");
+ FILE* varobjects;
+ varobjects=fopen("varobjs.txt","wr");
+ FILE* lower;
+ lower=fopen("lowerbounds.txt","wr");
- // SCIP_CALL(SCIPstopClock(relaxscip,varslottime));
- // SCIPinfoMessage(relaxscip, NULL, "Solving Time (sec) : %6f\n", SCIPgetClockTime(relaxscip, varslottime));
-
- /*add all the duals that are added onto the variabls under sumofaddonvars*/
- SCIP_Real sumofaddonvars=0;
- SCIP_Real percent;
- SCIP_Real percent2;
- int counter=0;
- SCIP_Real sumofsqsubgrad=0;
+ int maxiter=125;
+ fprintf(lower,"%d\n",maxiter);
-
- stepsize;
- /*Now we can start with our interation to maximize the dualmultipliers while minimizing the reformulated SCIP instance*/
- for(int k = 0; k < niters; ++k)
- {
- stepsize = 0.1;
- printf("%f\n\n",stepsize);
- sumofaddonvars = 0;
- SCIP_CALL(SCIPfreeTransform(relaxscip));
- /*******************************************************************************************************/
- /* The reformulation of the problem can be written as follows */
- //*>>>>>>>>>>>>>>>>>> min sum { (w[i]+sum{dual[j]})}x[i]-sum{dual[r]} <<<<<<<<<<<< */
- /*where i is nvars, j is nVarSlotConss, and r is nSlotConss for our case *******************************/
- /*******************************************************************************************************/
- SCIPvarchangeDuals(relaxscip,&vars,dualmultipliers,origobj);
+ for(int iter=1;iter<=maxiter;iter++)
+ {
+
+ for(int v=0;v<nvars;v++)
+ {
+ SCIP_VAR* var = vars[v];
+ double sum =SCIPvarGetObj(var);
+
+ vardata=SCIPvarGetData(var);
+ int* varids = SCIPvardataGetvarids(vardata);
+ int NVarInBadConss = SCIPvardataGetNVarInBadConss(vardata);
+
+ // printf("\n");
+ for(int t=0;t<NVarInBadConss;t++)
+ {
+ // printf("sum = %f, varid %d, dual %f, ", sum, varids[t],dualmultipliers[varids[t]]);
+ sum += dualmultipliers[varids[t]];
+ // fprintf(varobjects,"{%d, %f, %f\t",varids[t], dualmultipliers[varids[t]],sum);
+ }
+ // fprintf(varobjects,"}\n\n");
+ SCIP_CALL(SCIPaddVarObj(relaxscip,var,sum));
+ // if(sum>weights[v]){printf("new weight %f",SCIPvarGetObj(var));}
+
+ }
+ // printf("weight for v1 %f \t:= conss",solvals[1]);
+ // for(int s=0; s<listnconsvars[0];++s)
+ // {
+ // int id = listconsvarids[s];
+
+ // printf("(%s, duals = %f) \t",SCIPconsGetName(SCIPgetConss(scip)[id]), dualmultipliers[id]);
+ // }
+
+ SCIPinfoMessage(relaxscip, TimeCollector, "\n finished changing the variable's weight after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+
SCIP_CALL(SCIPaddOrigObjoffset(relaxscip,-1*sumofduals));
- // printf("*****************************upperbound*********************---> %f",SCIPgetUpperbound(relaxscip));
- // SCIP_CALL(SCIPprintOrigProblem(relaxscip, file, "lp", FALSE));
+ // SCIP_CALL(SCIPprintOrigProblem(relaxscip, AfterPreProcessing, "lp", FALSE));
+ SCIPsetMessagehdlrQuiet(relaxscip, TRUE);
+ // fclose(AfterPreProcessing);
+
+ SCIP_CALL( SCIPtransformProb(relaxscip) );
+ SCIP_CALL( SCIPsolve(relaxscip) );
+ relaxval = SCIPgetPrimalbound(relaxscip);
+ // printf("\ndualbound %f, primalbound %f \n",SCIPgetDualbound(relaxscip),SCIPgetPrimalbound(relaxscip));
+ SCIPdebugMessage("relaxation bound = %e status = %d\n", relaxval, SCIPgetStatus(relaxscip));
+ /*get the best solution*/
+ SCIP_SOL* bestsol = SCIPgetBestSol(relaxscip) ;
+ SCIP_SOL** sols = SCIPgetSols(relaxscip);
+ int nsols = SCIPgetNSols(relaxscip);
+
+ SCIP_Real* solvals;
+ SCIP_CALL(SCIPallocBufferArray(relaxscip,&solvals,nvars+1));
+
+
+ /*text output*/
+ SCIPinfoMessage(relaxscip, TimeCollector, "\n first iteration: problem solved after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ fprintf(solutions,"number of solutions %d, first iteration \t bound=%f, \t objsol=%f \n",nsols, SCIPgetPrimalbound(relaxscip),relaxval);
+ // SCIP_CALL(SCIPprintBestSol(relaxscip,solutions,FALSE));
+
+ /*store the solution in solvals so we can later export it to subgradient function*/
+ SCIP_Real lowerbound=0;
+ SCIPgetSolVals(relaxscip,bestsol,nvars,vars,solvals);
+ SCIP_CALL(SCIPprintSol(relaxscip,bestsol,dual,FALSE));
+
+ SCIP_Real compare=0;
+ for (int v = 0; v<nvars; ++v)
+ {
+ compare += solvals[v]*weights[v];
+ }
+
+ printf("compare value %f\n",compare);
+ // for(int s=0;s<nsols;s++)
+ // {
+ // SCIPgetSolVals(relaxscip,sols[s],nvars,vars,solvals);
+ // SCIP_CALL(SCIPprintSol(relaxscip,sols[s],dual,FALSE));
+ // SCIP_Real compare=0;
+ // for (int v = 0; v<nvars; ++v)
+ // {
+ // compare += solvals[v]*weights[v];
+ // }
+
+ // printf("compare value %f\n",compare);
+ // if(compare>lowerbound){lowerbound==compare;}
+ // }
+ // fprintf(dual,"now comes the biggest one\n");
+
+ // for(int s=0;s<nsols;s++)
+ // {
+ // SCIPgetSolVals(relaxscip,sols[s],nvars,vars,solvals);
+ // SCIP_CALL(SCIPprintSol(relaxscip,sols[s],dual,FALSE));
+ // SCIP_Real compare=0;
+ // for (int v = 0; v<nvars; ++v)
+ // {
+ // compare += solvals[v]*weights[v];
+ // }
+ // if(compare==lowerbound){break;}
+ // }
+
- SCIPsolveiteration(relaxscip,nSlotConss,&subgradients,2,&stepsize,&bestsolvals,&upperbound, k);
- //prinf("\t\t\t\t%fUpperbound \n",upperbound);
-
- SCIP_CALL(SCIPduplicateBlockMemoryArray(scip, &(relaxdata->bestsolvals), bestsolvals, nvars+1));
- // printf("\n best obj = %f \n",bestsolvals[nvars]);
+ // stepsize = 15/double(iter+1);
+ // fprintf(solutions, "\niteration %d\n",iter);
+ // fprintf(dual, "\niteration %d\n",iter);
+ // fprintf(variableinfo, "\niteration %d\n",iter);
+ // fprintf(varobjects, "\niteration %d\n",iter);
SCIP_CALL(SCIPaddOrigObjoffset(relaxscip,sumofduals));
-
- sumofduals = 0;
- for ( int r = 0; r < nSlotConss; ++r)
- {
- SCIP_CONS* cons = conss[r];
- // printf("--subgrad %f--",subgradients[r]);
- SCIP_Real subgradient = subgradients[r];
- sumofsqsubgrad += subgradients[r]*subgradients[r];
- }
-
- for ( int r = 0; r < nSlotConss; ++r)
- {
- SCIP_CONS* cons = conss[r];
- // printf("--subgrad %f--",subgradients[r]);
- SCIP_Real subgradient = subgradients[r];
- dualmultipliers[r]=SCIPconsGetMultiplier(relaxscip,&cons,subgradient,C,stepsize,false,dualmultipliers[r]); //returns the minumum
- sumofduals+=dualmultipliers[r]; //adds the negative of the minimum in each iteration
- //prinf("(%s,%f, sumofduals %f) \n",SCIPconsGetName(cons), dualmultipliers[r],sumofduals);
+ // SCIP_CALL( SCIPfreeTransform(relaxscip) );
+ // SCIP_CALL( SCIPtransformProb(relaxscip) );
+
+ counter = 0;
+ int checker = 0;
+ for(int r=0; r<nSlotConss;++r)
+ {
+ id = badconss[r];
+ double ax=-1;
+ for(int s=counter;s<(counter+listnconsvars[r]);++s)
+ {
+ // printf("%s->",SCIPvarGetName(vars[listconsvarids[s]]));
+ ax+=SCIPgetSolVal(relaxscip,bestsol,vars[listconsvarids[s]]);
+ // fprintf(subgrad,"%s\t,%f\t, sum %f",SCIPvarGetName(vars[listconsvarids[s]]),SCIPgetSolVal(relaxscip,bestsol,vars[listconsvarids[s]]),ax);
+
+ }
+
+ counter += listnconsvars[r];
+ if(ax>0){checker++;}
+ subgradients[r]=ax;
+ // fprintf(subgrad, "\n subgrad = %f \t",subgradients[r]);
+
}
+ if(checker==0){printf("#*#*#*result found\n"); break;}
-
+ SCIP_CALL( SCIPfreeTransform(relaxscip) );
+ SCIP_CALL( SCIPtransformProb(relaxscip) );
+
- ++stoppingiter;
- if(bestsolvals[nvars]!=bestsolvals[nvars+k+1]){++counter;}
- else{counter=0;}
- if(counter==10){ break; }
- }
+
+ for (int v = 0; v<nvars; ++v)
+ {
+ SCIP_VAR* var = vars[v];
+
+ SCIP_CALL(SCIPchgVarObj(relaxscip,var,weights[v]));
+ // fprintf(variableinfo,"(%s,%f,%f)->%f\n",SCIPvarGetName(var),solvals[v],SCIPvarGetObj(var), weights[v]);
+ lowerbound += solvals[v]*weights[v];
+ }
+ fprintf(dual,"dualbound = %f, lowerbound=%f, norm of subgrad %f\t",SCIPgetPrimalbound(relaxscip),lowerbound, getnorm(subgradients,nSlotConss,stepsize));
+ fprintf(lower,"%f\n",lowerbound);
- SCIP_CALL(SCIPstopClock(relaxscip,varslottime));
- SCIPinfoMessage(relaxscip, NULL, "Solving Time (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ // stepsize = (SCIPgetPrimalbound(relaxscip)-lowerbound)/getnorm(subgradients,nSlotConss,stepsize);
+ SCIP_CALL( SCIPfreeTransform(relaxscip) );
+ fprintf(solutions, "lowerbound = %f \n ", lowerbound);
+ SCIPinfoMessage(relaxscip, TimeCollector, "\n subgradients found after (sec) : %f\n, lowerbound = %f \n", SCIPgetClockTime(relaxscip, varslottime),lowerbound);
+
+ //add back the sum of the duals we subtracted from the main obj function
- FILE* saveniter;
- saveniter = fopen("values.txt","w");
- fprintf(saveniter,"%d\n",stoppingiter);
- fclose(saveniter);
+ int sum=0;
+ sumofduals = 0;
- FILE* savebestsols;
- for(int v=nvars+1; v<nvars+1+stoppingiter; ++v)
- {
- // prinf("(%s bsv= %f\n", SCIPvarGetName(vars[v]),bestsolvals[v]);
- savebestsols = fopen("values.txt","a");
- fprintf(savebestsols,"%f\n", bestsolvals[v]);
- // fputs("This is testing for fputs...\n",were);
- fclose(savebestsols);
+ for(int r=0; r<nSlotConss;++r)
+ {
+ dualmultipliers[r] += subgradients[r]*stepsize;
+ if(dualmultipliers[r]<0){dualmultipliers[r]=0;}
+
+ sum+=dualmultipliers[r];
+ // fprintf(dual," then %f step size %f \n",dualmultipliers[r], stepsize);
+ }
+ sumofduals=sum;
+ // fprintf(dual,"iteration %d, sumofduals=%f\n",iter, sumofduals);
+ SCIPinfoMessage(relaxscip, TimeCollector, "\n new dual found after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ // if(checker==0){printf("solution found in %d iterations\n",iter); break;}
}
+ SCIPfreeTransform(relaxscip);
+ fclose(variableinfo);
+ fclose(dual);
+ fclose(subgrad);
+ fclose(varobjects);
+ fclose(solutions);
+ fclose(lower);
- for(int v=0; v<nvars; ++v)
- {
- SCIP_VAR* var = vars[v];
- // printf("orobj %f|",origobj[v]);
- // vardata = SCIPvarGetData(var);
- // SCIP_CALL(SCIPVaraddDualMultiplier(relaxscip,&var,dualmultipliers));
- // printf("%f",SCIPvarGetQuotient(vardata));
- // vardataDelete(relaxscip,&vardata);
- }
-
-
+ /* free memory */
+ SCIPhashmapFree(&varmap);
+ SCIP_CALL( SCIPfree(&relaxscip) );
-
return SCIP_OKAY;
}
@@ -381,17 +546,19 @@ static
SCIP_DECL_RELAXEXEC(relaxExeclagr)
{
/*lint --e{715}*/
-SCIP* relaxscip;
+ SCIP* relaxscip;
SCIP_HASHMAP* varmap;
SCIP_HASHMAP* consmap;
SCIP_CONS** conss;
SCIP_PROBDATA* probdata;
SCIP_VARDATA* vardata;
+
SCIP_Real relaxval;
SCIP_Bool valid;
int nconss;
int i;
- int c;
+ int counter;
+ int id;
*lowerbound = -SCIPinfinity(scip);
@@ -400,28 +567,6 @@ SCIP* relaxscip;
/* we can only run if none of the present constraints expect their variables to be binary or integer during transformation */
conss = SCIPgetConss(scip);
nconss = SCIPgetNConss(scip);
-
-
- for( c = 0; c < nconss; ++c )
- {
- const char* conshdlrname;
-
- conshdlrname = SCIPconshdlrGetName(SCIPconsGetHdlr(conss[c]));
-
- /* skip if there are any "and", "linking", or", "orbitope", "pseudoboolean", "superindicator", "xor" or new/unknown constraints */
- if( strcmp(conshdlrname, "SOS1") != 0 && strcmp(conshdlrname, "SOS2") != 0 && strcmp(conshdlrname, "abspower") != 0
- && strcmp(conshdlrname, "bivariate") != 0 && strcmp(conshdlrname, "bounddisjunction") != 0
- && strcmp(conshdlrname, "cardinality") != 0 && strcmp(conshdlrname, "components") != 0
- && strcmp(conshdlrname, "conjunction") != 0 && strcmp(conshdlrname, "countsols") != 0
- && strcmp(conshdlrname, "cumulative") != 0 && strcmp(conshdlrname, "disjunction") != 0
- && strcmp(conshdlrname, "indicator") != 0 && strcmp(conshdlrname, "integral") != 0
- && strcmp(conshdlrname, "knapsack") != 0 && strcmp(conshdlrname, "linear") != 0
- && strcmp(conshdlrname, "logicor") != 0 && strcmp(conshdlrname, "nonlinear") != 0
- && strcmp(conshdlrname, "orbisack") != 0 && strcmp(conshdlrname, "quadratic") != 0
- && strcmp(conshdlrname, "setppc") != 0 && strcmp(conshdlrname, "soc") != 0
- && strcmp(conshdlrname, "symresack") != 0 && strcmp(conshdlrname, "varbound") != 0 )
- return SCIP_OKAY;
- }
/* create the variable mapping hash map */
SCIP_CALL( SCIPcreate(&relaxscip) );
@@ -429,178 +574,285 @@ SCIP* relaxscip;
valid = FALSE;
SCIP_CALL( SCIPcopy(scip, relaxscip, varmap, consmap, "relaxscip", FALSE, FALSE, FALSE, FALSE, &valid) );
- /**************************************************************************************************************/
- /*First, */
- //*the probdata: where we get to identify the bad constraint we want to formulate(in our case, the slot conss) */
- /***************************************************************************************************************/
- SCIPcreateprobdata(relaxscip,&probdata); /*will be used to identify the # of slot(bad) constraints*/
- int nSlotConss = SCIPgetNSlotConss(probdata); //number of bad(slot) constraint
- int* badconss;
- SCIP_CALL(SCIPallocBufferArray(relaxscip, &badconss, nSlotConss)); //the badconss array will contain the row number of the bad conss
- int counter = 0; //counter will increase based the condition of inequality withn row.
- for (int t=0; t<nconss; t++) /* (3) */
- {
- SCIP_CONS* cons = SCIPgetConss(relaxscip)[t];
- if(SCIPconsGetLhs(relaxscip,cons,&valid)==-SCIPinfinity(relaxscip)) /*<We get the slot constraints based on the inquality*/
- {
- badconss[counter]=t;
- printf("(%d,%d)--%s\t",counter, t,SCIPconsGetName(cons));
- counter++;
- }
- }
-
- int nvars = SCIPgetNVars(relaxscip);
- SCIP_VAR** vars = SCIPgetVars(relaxscip);
-
- int* consids;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&consids,nSlotConss));
-
- /* we then create the vardata function for each variable, to see at which constraint the variable is found*/
- SCIP_CLOCK* varslottime; //to help us record the time
- SCIP_CALL( SCIPcreateClock(relaxscip, &varslottime) ); //* start time counting*
- SCIP_CALL(SCIPstartClock(relaxscip,varslottime));
- int maxnconsvars=0;
- int nconsvars=0;
+ // /**************************************************************************************************************/
+ // /*First, */
+ // //*the probdata: where we get to identify the bad constraint we want to formulate(in our case, the slot conss) */
+ // /***************************************************************************************************************/
+ // int nvars = SCIPgetNVars(relaxscip);
+ // SCIP_VAR** vars = SCIPgetVars(relaxscip);
+ // SCIP_VAR** varbuffers;
+ // int* badconss;
-
- int id;
- for(int r=0;r<nSlotConss;r++)
- {
- id = badconss[r]; //gets the index for the bad constraints
- SCIP_CONS* cons = SCIPgetConss(relaxscip)[id]; //gets teh bad constraints
- SCIP_CALL(SCIPgetConsNVars(relaxscip,cons,&nconsvars,&valid));
- if(maxnconsvars<nconsvars){maxnconsvars=nconsvars;} //get the max number of variables found at each constraint
- consids[r]=0;
- // printf("%s\n",SCIPconsGetName(cons));
- }
-
- SCIP_VAR** varbuffers;
- SCIP_CALL(SCIPallocBufferArray(relaxscip, &varbuffers, maxnconsvars));
- for (int v = 0; v < maxnconsvars; v++)
- {
- varbuffers[v] =vars[v];
- }
-
- SCIP_Real* origobj;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&origobj,nvars));
- for (int v = 0; v < nvars; v++)
- {
- int* varid;
- int nVarSlotConss=0;
- nconsvars = 0;
- SCIP_VAR* var = vars[v];
- origobj[v]= SCIPvarGetObj(var);
-
- int varindex = SCIPvarGetIndex(var); /* (2) */
- assert(varindex!= NULL);
-
- // printf("%s****%d\n",SCIPvarGetName(var),varindex);
- for (int r = 0; r < nSlotConss; ++r)
- {
- id = badconss[r];
- SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
- // printf("%s \t",SCIPconsGetName(cons));
- SCIP_CALL(SCIPgetConsNVars(relaxscip, cons, &nconsvars, &valid));
- SCIP_CALL(SCIPgetConsVars(relaxscip, cons, varbuffers, nconsvars, &valid));
- if (!valid){
- abort(); }
-
- for (int j = 0; j < nconsvars; ++j) /* (8) */
- {
- SCIP_VAR* varx = varbuffers[j];
- int varbufindex = SCIPvarGetIndex(varx);
- assert(varbufindex != NULL);
- // printf("%s\t \t%d",SCIPvarGetName(varx),varbufindex);
+ // SCIPcreateprobdata(relaxscip,&probdata,SCIPgetConss(relaxscip),vars,&varbuffers,&badconss); /*will be used to identify the # of slot(bad) constraints*/
+ // int nSlotConss = SCIPgetNSlotConss(probdata); //number of bad(slot) constraint
+ // int allnconsvars = SCIPgetallnconsvars(probdata); //sum of all nconsvars, used for creating later on an array to collect the list of varids in each row
+ // int* listnconsvars = SCIPlistnconsvars(probdata);
+ // int* listconsvarids = SCIPlistconsvarids(probdata);
+
+ // /* we then create the vardata function for each variable, to see at which constraint the variable is found*/
+ // FILE* TimeCollector;
+ // TimeCollector = fopen("time.txt","w");
+ // SCIP_CLOCK* varslottime; //to help us record the time
+ // SCIP_CALL( SCIPcreateClock(relaxscip, &varslottime) ); //* start time counting*
+ // SCIP_CALL(SCIPstartClock(relaxscip,varslottime));
+
+ // // int nconsvars=0;
+ // int* consids;
+
+ // SCIP_Real* weights;
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip,&weights,nvars));
+
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip,&consids,nSlotConss));
+
+ // for (int v = 0; v < nvars; v++)
+ // {
+ // SCIP_VAR* var = vars[v];
+ // weights[v]=SCIPvarGetObj(var);
+ // }
+
+ // for (int v = 0; v < nvars; v++)
+ // {
+ // int* varids;
+ // int NVarInBadConss=0;
+ // int nconsvars = 0;
+ // SCIP_VAR* var = vars[v];
+
+ // int varindex = SCIPvarGetIndex(var); /* (2) */
+ // assert(varindex!= NULL);
+
+ // // printf("%s****%d\n",SCIPvarGetName(var),varindex);
+ // for (int r = 0; r < nSlotConss; ++r)
+ // {
+ // id = badconss[r];
+ // SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // // printf("%s \t",SCIPconsGetName(cons));
+ // SCIP_CALL(SCIPgetConsNVars(relaxscip, cons, &nconsvars, &valid));
+ // SCIP_CALL(SCIPgetConsVars(relaxscip, cons, varbuffers, nconsvars, &valid));
+ // if (!valid){
+ // abort(); }
+
+ // for (int j = 0; j < nconsvars; ++j) /* (8) */
+ // {
+ // SCIP_VAR* varx = varbuffers[j];
+ // int varbufindex = SCIPvarGetIndex(varx);
+ // assert(varbufindex != NULL);
+ // // printf("%s\t \t%d",SCIPvarGetName(varx),varbufindex);
- /** if var[i] is in cons[c], write conspointer in VarConss and increase nVarConsscounter */
- if (varindex == varbufindex) /* (9) */
- {
+ // /** if var[i] is in cons[c], write conspointer in VarConss and increase nVarConsscounter */
+ // if (varindex == varbufindex) /* (9) */
+ // {
- // VarSlotConss[nVarSlotConss] = cons;
- consids[nVarSlotConss]=id;
- nVarSlotConss++;
- // printf(" %s \t,",SCIPconsGetName(cons));
- }
- }
- }
- SCIP_CALL(SCIPallocBufferArray(relaxscip, &varid, nVarSlotConss));
- for(int t=0;t<nVarSlotConss;t++)
- {
- varid[t]=consids[t];
- // printf("%d \t",varid[t]);
- }
+ // // VarSlotConss[NVarInBadConss] = cons;
+ // consids[NVarInBadConss]=id;
+ // NVarInBadConss++;
+ // // printf(" %s \t,",SCIPconsGetName(cons));
+ // }
+ // }
+ // }
+
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip, &varids, NVarInBadConss));
+ // for(int t=0;t<NVarInBadConss;t++)
+ // {
+ // varids[t]=consids[t];
+ // // printf("%d \t",varids[t]);
+ // }
+
+ // // vardata=SCIPvarGetData(var);
+ // SCIP_CALL(SCIPallocBlockMemory(scip , &vardata));
+ // SCIP_CALL(SCIPduplicateBlockMemoryArray(scip, &(vardata->varids), varids, NVarInBadConss));
+ // vardata->NVarInBadConss = NVarInBadConss; /**copy nVarConss to VarData */
+ // vardata->varids = varids;
+ // // /**set the variable data to the variable*/
+ // SCIPvarSetData(var,vardata);
+ // }
+
+ // // SCIP_CALL(SCIPstopClock(relaxscip,varslottime));
+
- // vardata=SCIPvarGetData(var);
- SCIP_CALL(SCIPallocBlockMemory(scip , &vardata));
- SCIP_CALL(SCIPduplicateBlockMemoryArray(scip, &(vardata->varid), varid, nVarSlotConss));
- vardata->nVarSlotConss = nVarSlotConss; /**copy nVarConss to VarData */
- // // vardata->varquotient = varquotient;
- vardata->varid = varid;
- // /**set the variable data to the variable*/
- SCIPvarSetData(var,vardata);
- }
+ // FILE* AfterPreProcessing;
+ // AfterPreProcessing = fopen("AfterPreProcessing.txt","w+");
+
+ // SCIP_CALL(SCIPprintOrigProblem(relaxscip, AfterPreProcessing, "lp", FALSE));
+
+ // SCIPinfoMessage(relaxscip, TimeCollector, "\n row and column identified in (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ // for(int r=0;r<nSlotConss;r++)
+ // {
+ // id = badconss[r];
+ // SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // SCIP_CALL(SCIPdelCons(relaxscip,cons));
+ // }
+
+ // /******************************************************************************************************************/
+ // /*Next, we will do the initial iteration of finding the dual mulpliers of each slot conss, and their sum(dualsum) */
+ // /* In the end, we will subtract this sum from the objective of the function. */
+ // /* It's initial, because while we would search for more dual multipliers to solve the Lagrangian relaxation */
+ // /******************************************************************************************************************/
+ // SCIP_Real* dualmultipliers;
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip,&dualmultipliers,nSlotConss));
+
+ // SCIP_Real* subgradients;
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip,&subgradients,nSlotConss));
+ // //initialize subgradients;
+ // SCIP_Real stepsize = 1.00000;
+ // SCIP_Real sumofduals=0;
+ // for ( int r = 0; r < nSlotConss; ++r)
+ // {
+ // // id = badconss[r];
+ // // SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // //if k=1 iteration//
+ // dualmultipliers[r] = 1;
+ // sumofduals+=dualmultipliers[r]; //adds the negative of the minimum in each iteration
+
+ // }
- // SCIP_CALL(SCIPstopClock(relaxscip,varslottime));
- SCIPinfoMessage(relaxscip, NULL, "\n row and column identified in (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
- /******************************************************************************************************************/
- /*Next, we will do the initial iteration of finding the dual mulpliers of each slot conss, and their sum(dualsum) */
- /* In the end, we will subtract this sum from the objective of the function. */
- /* It's initial, because while we would search for more dual multipliers to solve the Lagrangian relaxation */
- /******************************************************************************************************************/
- SCIP_Real* dualmultipliers;
- SCIP_CALL(SCIPallocBufferArray(relaxscip,&dualmultipliers,nSlotConss));
- int sumofduals=0;
- for ( int r = 0; r < nSlotConss; ++r)
- {
- id = badconss[r];
- SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // /*******************************************************************************************************/
+ // /* The reformulation of the problem can be written as follows */
+ // //*>>>>>>>>>>>>>>>>>> min sum { (w[i]+sum{dual[j]})}x[i]-sum{dual[r]} <<<<<<<<<<<< */
+ // /*where i is nvars, j is NVarInBadConss, and r is nSlotConss for our case *******************************/
+ // /****************************************************************************************************************/
+ // /* The following function will add the following to the obj(weight) of the variable, */
+ // //* the obj(weight) of var + the sum of the dualmultipliers of bad constraints which contains this variable */
+ // /****************************************************************************************************************/
+
+
+ // FILE* solutions;
+ // solutions = fopen("sol.txt","wr");
+ // FILE* dual;
+ // dual= fopen("dual.txt","wr");
+ // FILE* variableinfo;
+ // variableinfo = fopen("var.txt","wr");
+ // FILE* subgrad;
+ // subgrad = fopen("subgrads.txt","wr");
+ // FILE* varobjects;
+ // varobjects=fopen("varobjs.txt","wr");
+
+ // int maxiter=20;
+ // SCIP_Real lagrdual =-SCIPinfinity(scip);
+
+ // for(int iter=1;iter<=maxiter;iter++)
+ // {
-
- // dualmultipliers[r]=SCIPconsGetMultiplier(relaxscip,&cons, 0,0,0,true,0); //returns the minumum
- dualmultipliers[r] = 0;
- sumofduals+=dualmultipliers[r]; //adds the negative of the minimum in each iteration
+ // for(int v=0;v<nvars;v++)
+ // {
+ // SCIP_VAR* var = vars[v];
+ // double sum =SCIPvarGetObj(var);
+
+ // vardata=SCIPvarGetData(var);
+ // int* varids = SCIPvardataGetvarids(vardata);
+ // int NVarInBadConss = SCIPvardataGetNVarInBadConss(vardata);
+
+ // for(int t=0;t<NVarInBadConss;t++)
+ // {
+ // sum += dualmultipliers[varids[t]];
+ // fprintf(varobjects,"{%d, %f, %f\t",varids[t], dualmultipliers[varids[t]],sum);
+ // }
+ // fprintf(varobjects,"}\n\n");
+ // SCIP_CALL(SCIPaddVarObj(relaxscip,var,sum));
+
+ // }
- }
-
- /*******************************************************************************************************/
- /* The reformulation of the problem can be written as follows */
- //*>>>>>>>>>>>>>>>>>> min sum { (w[i]+sum{dual[j]})}x[i]-sum{dual[r]} <<<<<<<<<<<< */
- /*where i is nvars, j is nVarSlotConss, and r is nSlotConss for our case *******************************/
- /****************************************************************************************************************/
- /* The following function will add the following to the obj(weight) of the variable, */
- //* the obj(weight) of var + the sum of the dualmultipliers of bad constraints which contains this variable */
- /****************************************************************************************************************/
-
-
+ // SCIPinfoMessage(relaxscip, TimeCollector, "\n finished changing the variable's weight after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+
+ // SCIP_CALL(SCIPaddOrigObjoffset(relaxscip,-1*sumofduals));
+ // SCIP_CALL(SCIPprintOrigProblem(relaxscip, AfterPreProcessing, "lp", FALSE));
+ // SCIPsetMessagehdlrQuiet(relaxscip, TRUE);
+ // // fclose(AfterPreProcessing);
+
+ // SCIP_CALL( SCIPtransformProb(relaxscip) );
+ // SCIP_CALL( SCIPsolve(relaxscip) );
+ // relaxval = SCIPgetPrimalbound(relaxscip);
+ // SCIPdebugMessage("relaxation bound = %e status = %d\n", relaxval, SCIPgetStatus(relaxscip));
+ // /*get the best solution*/
+ // SCIP_SOL* bestsol = SCIPgetBestSol(relaxscip) ;
+ // SCIP_Real* solvals;
+ // SCIP_CALL(SCIPallocBufferArray(relaxscip,&solvals,nvars+1));
+
- /* change variable types */
- for( i = 0; i < SCIPgetNVars(relaxscip); ++i )
- {
- SCIP_VAR* var;
- SCIP_Bool infeasible;
+ // /*text output*/
+ // SCIPinfoMessage(relaxscip, TimeCollector, "\n first iteration: problem solved after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ // fprintf(solutions,"first iteration \t bound=%f, \t objsol=%f \n",SCIPgetPrimalbound(relaxscip),relaxval);
+ // SCIP_CALL(SCIPprintBestSol(relaxscip,solutions,FALSE));
- var = SCIPgetVars(relaxscip)[i];
- assert(var != NULL);
+ // /*store the solution in solvals so we can later export it to subgradient function*/
+ // SCIPgetSolVals(relaxscip,bestsol,nvars,vars,solvals);
+
- SCIP_CALL( SCIPchgVarType(relaxscip, var, SCIP_VARTYPE_CONTINUOUS, &infeasible) );
- assert(!infeasible);
- }
+ // stepsize = 1/double(iter+1);
+ // fprintf(solutions, "\niteration %d\n",iter);
+ // // fprintf(dual, "\niteration %d\n",iter);
+ // fprintf(variableinfo, "\niteration %d\n",iter);
+ // fprintf(varobjects, "\niteration %d\n",iter);
+
+ // SCIP_CALL(SCIPaddOrigObjoffset(relaxscip,sumofduals));
+ // // SCIP_CALL( SCIPfreeTransform(relaxscip) );
+ // // SCIP_CALL( SCIPtransformProb(relaxscip) );
+
+ // counter = 0;
+ // int checker = 0;
+ // for(int r=0; r<nSlotConss;++r)
+ // {
+ // id = badconss[r];
+
+ // SCIP_CONS* cons = SCIPgetConss(relaxscip)[id];
+ // // SCIP_CALL(SCIPgetConsNVars(relaxscip, cons, &nconsvars, &valid));
+
+ // double ax=-1;
+
+ // for(int s=counter;s<(counter+listnconsvars[r]);++s)
+ // {
+ // // printf("%s->",SCIPvarGetName(vars[listconsvarids[s]]));
+ // ax+=SCIPgetSolVal(relaxscip,bestsol,vars[listconsvarids[s]]);
+ // fprintf(subgrad,"%s\t,%f\t, sum %f",SCIPvarGetName(vars[listconsvarids[s]]),SCIPgetSolVal(relaxscip,bestsol,vars[listconsvarids[s]]),ax);
+
+ // }
+
+ // counter += listnconsvars[r];
+ // if(ax<0){subgradients[r]==0; }
+ // else{subgradients[r]==ax;checker++;}
+
+ // }
+
+
+
- SCIP_CALL( SCIPtransformProb(relaxscip) );
- FILE* saveniter;
- saveniter = fopen("time.txt","w");
+ // SCIP_CALL( SCIPfreeTransform(relaxscip) );
+ // SCIP_CALL( SCIPtransformProb(relaxscip) );
+ // for (int v = 0; v<nvars; ++v)
+ // {
+ // SCIP_VAR* var = vars[v];
+ // SCIP_CALL(SCIPchgVarObj(relaxscip,var,weights[v]));
+ // fprintf(variableinfo,"(%s,%f,%f)->%f\n",SCIPvarGetName(var),solvals[v],SCIPvarGetObj(var), weights[v]);
+ // }
+ // SCIP_CALL( SCIPfreeTransform(relaxscip) );
+ // SCIPinfoMessage(relaxscip, TimeCollector, "\n subgradients found after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+
- SCIP_CALL(SCIPprintOrigProblem(relaxscip, saveniter, "lp", FALSE));
- SCIPsetMessagehdlrQuiet(relaxscip, TRUE);
- fclose(saveniter);
+ // //add back the sum of the duals we subtracted from the main obj function
+
+ // int sum=0;
+ // for(int r=0; r<nSlotConss;++r)
+ // {
+ // dualmultipliers[r] += subgradients[r]+stepsize;
+ // if(dualmultipliers[r]<0){dualmultipliers[r]=0;}
+
+ // sum+=dualmultipliers[r];
+ // // fprintf(dual," then %f \n",dualmultipliers[r]);
+ // }
+ // sumofduals=sum;
+ // fprintf(dual,"iteration %d, sumofduals=%f\n",iter, sumofduals);
+ // SCIPinfoMessage(relaxscip, TimeCollector, "\n new dual found after (sec) : %f\n", SCIPgetClockTime(relaxscip, varslottime));
+ // if(checker==0){printf("solution found in %d iterations\n",iter); break;}
+ // }
+ // SCIPfreeTransform(relaxscip);
+ // fclose(variableinfo);
+ // fclose(dual);
+ // fclose(subgrad);
+ // fclose(varobjects);
- SCIP_CALL( SCIPsolve(relaxscip) );
- relaxval = SCIPgetPrimalbound(relaxscip);
- SCIPdebugMessage("relaxation bound = %e status = %d\n", relaxval, SCIPgetStatus(relaxscip));
if( SCIPgetStatus(relaxscip) == SCIP_STATUS_OPTIMAL )
{
@@ -608,6 +860,8 @@ SCIP* relaxscip;
if( (! SCIPisRelaxSolValid(scip)) || SCIPisGT(scip, relaxval, SCIPgetRelaxSolObj(scip)) )
{
SCIPdebugMsg(scip, "Setting LP relaxation solution, which improved upon earlier solution\n");
+
+
SCIP_CALL( SCIPclearRelaxSolVals(scip, relax) );
for( i = 0; i < SCIPgetNVars(scip); ++i )
@@ -628,6 +882,7 @@ SCIP* relaxscip;
}
SCIPdebugMsg(scip, "LP lower bound = %g\n", relaxval);
+
*lowerbound = relaxval;
*result = SCIP_SUCCESS;
}
diff --git a/src/relax_lagr.h b/src/relax_lagr.h
index 44a1f2691e841d8524f9509ea2caf8c6483d7429..352b5fd8fdb275e97a5f12298d05511f518fec26 100644
--- a/src/relax_lagr.h
+++ b/src/relax_lagr.h
@@ -39,6 +39,15 @@ SCIP_RETCODE SCIPincludeRelaxlagrangian(
#ifdef __cplusplus
}
+
+int SCIPvardataGetNVarInBadConss(
+ SCIP_VARDATA* vardata /**< variable data */
+);
+
+int* SCIPvardataGetvarids(
+ SCIP_VARDATA* vardata /**< variable data */
+);
+
#endif
#endif
diff --git a/src/relax_lagr32.cpp b/src/relax_lagr32.cpp
index 646dc85e19b9fe0f40298c5c12a31bd27984e1d7..61f4b2358b5c4c46c7258324113dd3e769ca7894 100644
--- a/src/relax_lagr32.cpp
+++ b/src/relax_lagr32.cpp
@@ -225,7 +225,7 @@ SCIP_DECL_RELAXINIT(relaxInitlagr)
for(int k = 0; k < niters; ++k)
{
- // stepsize = 10*(1/(0.5*(k+1)));
+ stepsize = 10*(1/(0.5*(k+1)));
// printf("%f\n\n",stepsize);
sumofaddonvars = 0;
SCIP_CALL(SCIPfreeTransform(relaxscip));
@@ -239,9 +239,7 @@ SCIP_DECL_RELAXINIT(relaxInitlagr)
// printf("*****************************upperbound*********************---> %f",SCIPgetUpperbound(relaxscip));
// SCIP_CALL(SCIPprintOrigProblem(relaxscip, file, "lp", FALSE));
- SCIPsolveiteration(relaxscip,nSlotConss,&subgradients,2,&stepsize,&bestsolvals,&upperbound, k);
- //prinf("\t\t\t\t%fUpperbound \n",upperbound);
-
+ SCIPsolveiteration(relaxscip,nSlotConss,&subgradients,2,&stepsize,&bestsolvals,&upperbound, k);
SCIP_CALL(SCIPduplicateBlockMemoryArray(scip, &(relaxdata->bestsolvals), bestsolvals, nvars+1));
// printf("\n best obj = %f \n",bestsolvals[nvars]);
diff --git a/subgrads.txt b/subgrads.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/time.txt b/time.txt
index 4e649a39a1f38c4ccbcb466bd687f4540a08c40b..b7aef1dabdbc8cdb0a811f651e446132db02f872 100644
--- a/time.txt
+++ b/time.txt
@@ -1,119 +1,1127 @@
-\ SCIP STATISTICS
-\ Problem name : data/6f4s.lp_relaxscip
-\ Variables : 54 (27 binary, 0 integer, 0 implicit integer, 27 continuous)
-\ Constraints : 36
-Minimize
- Obj: +120 t_x_{6}_{120} +105 t_x_{6}_{105} +90 t_x_{6}_{90} +75 t_x_{6}_{75} +60 t_x_{6}_{60} +45 t_x_{6}_{45}
- +30 t_x_{6}_{30} +15 t_x_{6}_{15} +120 t_x_{5}_{120} +105 t_x_{5}_{105} +90 t_x_{5}_{90} +75 t_x_{5}_{75}
- +60 t_x_{5}_{60} +45 t_x_{5}_{45} +30 t_x_{5}_{30} +15 t_x_{5}_{15} +120 t_x_{4}_{120} +105 t_x_{4}_{105}
- +90 t_x_{4}_{90} +75 t_x_{4}_{75} +60 t_x_{4}_{60} +45 t_x_{4}_{45} +30 t_x_{4}_{30} +15 t_x_{4}_{15}
- +105 t_x_{3}_{105} +90 t_x_{3}_{90} +75 t_x_{3}_{75} +60 t_x_{3}_{60} +45 t_x_{3}_{45} +30 t_x_{3}_{30}
- +15 t_x_{3}_{15} +120 t_x_{2}_{120} +105 t_x_{2}_{105} +90 t_x_{2}_{90} +75 t_x_{2}_{75} +60 t_x_{2}_{60}
- +45 t_x_{2}_{45} +30 t_x_{2}_{30} +15 t_x_{2}_{15} +120 t_x_{1}_{120} +105 t_x_{1}_{105} +90 t_x_{1}_{90}
- +75 t_x_{1}_{75} +60 t_x_{1}_{60} +45 t_x_{1}_{45} +30 t_x_{1}_{30} +15 t_x_{1}_{15} +1 t_x_{1}_{0}
- +120 t_x_{3}_{120}
-Subject to
- slot1: +1 t_x_{1}_{0} +1 t_x_{2}_{0} <= +1
- slot2: +1 t_x_{1}_{15} +1 t_x_{2}_{15} <= +1
- slot3: +1 t_x_{1}_{30} +1 t_x_{2}_{30} <= +1
- slot4: +1 t_x_{1}_{45} +1 t_x_{2}_{45} <= +1
- slot5: +1 t_x_{1}_{60} +1 t_x_{2}_{60} <= +1
- slot6: +1 t_x_{1}_{75} +1 t_x_{2}_{75} <= +1
- slot7: +1 t_x_{1}_{90} +1 t_x_{2}_{90} <= +1
- slot8: +1 t_x_{1}_{105} +1 t_x_{2}_{105} <= +1
- F_6: +1 t_x_{6}_{0} +1 t_x_{6}_{15} +1 t_x_{6}_{30} +1 t_x_{6}_{45} +1 t_x_{6}_{60} +1 t_x_{6}_{75} +1 t_x_{6}_{90}
- +1 t_x_{6}_{105} +1 t_x_{6}_{120} = +1
- slot10: +1 t_x_{2}_{0} +1 t_x_{3}_{0} +1 t_x_{4}_{0} <= +1
- slot11: +1 t_x_{2}_{15} +1 t_x_{3}_{15} +1 t_x_{4}_{15} <= +1
- slot12: +1 t_x_{2}_{30} +1 t_x_{3}_{30} +1 t_x_{4}_{30} <= +1
- slot13: +1 t_x_{2}_{45} +1 t_x_{3}_{45} +1 t_x_{4}_{45} <= +1
- slot14: +1 t_x_{2}_{60} +1 t_x_{3}_{60} +1 t_x_{4}_{60} <= +1
- slot15: +1 t_x_{2}_{75} +1 t_x_{3}_{75} +1 t_x_{4}_{75} <= +1
- slot16: +1 t_x_{2}_{90} +1 t_x_{3}_{90} +1 t_x_{4}_{90} <= +1
- F_4: +1 t_x_{4}_{0} +1 t_x_{4}_{15} +1 t_x_{4}_{30} +1 t_x_{4}_{45} +1 t_x_{4}_{60} +1 t_x_{4}_{75} +1 t_x_{4}_{90}
- +1 t_x_{4}_{105} +1 t_x_{4}_{120} = +1
- F_2: +1 t_x_{2}_{0} +1 t_x_{2}_{15} +1 t_x_{2}_{30} +1 t_x_{2}_{45} +1 t_x_{2}_{60} +1 t_x_{2}_{75} +1 t_x_{2}_{90}
- +1 t_x_{2}_{105} +1 t_x_{2}_{120} = +1
- slot19: +1 t_x_{1}_{0} +1 t_x_{3}_{0} +1 t_x_{5}_{0} <= +1
- slot20: +1 t_x_{1}_{15} +1 t_x_{3}_{15} +1 t_x_{5}_{15} <= +1
- slot21: +1 t_x_{1}_{30} +1 t_x_{3}_{30} +1 t_x_{5}_{30} <= +1
- slot22: +1 t_x_{1}_{45} +1 t_x_{3}_{45} +1 t_x_{5}_{45} <= +1
- slot23: +1 t_x_{1}_{60} +1 t_x_{3}_{60} +1 t_x_{5}_{60} <= +1
- slot24: +1 t_x_{1}_{75} +1 t_x_{3}_{75} +1 t_x_{5}_{75} <= +1
- F_1: +1 t_x_{1}_{0} +1 t_x_{1}_{15} +1 t_x_{1}_{30} +1 t_x_{1}_{45} +1 t_x_{1}_{60} +1 t_x_{1}_{75} +1 t_x_{1}_{90}
- +1 t_x_{1}_{105} +1 t_x_{1}_{120} = +1
- F_3: +1 t_x_{3}_{0} +1 t_x_{3}_{15} +1 t_x_{3}_{30} +1 t_x_{3}_{45} +1 t_x_{3}_{60} +1 t_x_{3}_{75} +1 t_x_{3}_{90}
- +1 t_x_{3}_{105} +1 t_x_{3}_{120} = +1
- F_5: +1 t_x_{5}_{0} +1 t_x_{5}_{15} +1 t_x_{5}_{30} +1 t_x_{5}_{45} +1 t_x_{5}_{60} +1 t_x_{5}_{75} +1 t_x_{5}_{90}
- +1 t_x_{5}_{105} +1 t_x_{5}_{120} = +1
- slot28: +1 t_x_{4}_{0} +1 t_x_{5}_{0} +1 t_x_{6}_{0} <= +1
- slot29: +1 t_x_{4}_{15} +1 t_x_{5}_{15} +1 t_x_{6}_{15} <= +1
- slot30: +1 t_x_{4}_{30} +1 t_x_{5}_{30} +1 t_x_{6}_{30} <= +1
- slot31: +1 t_x_{4}_{45} +1 t_x_{5}_{45} +1 t_x_{6}_{45} <= +1
- slot32: +1 t_x_{4}_{60} +1 t_x_{5}_{60} +1 t_x_{6}_{60} <= +1
- slot33: +1 t_x_{4}_{75} +1 t_x_{5}_{75} +1 t_x_{6}_{75} <= +1
- slot35: +1 t_x_{4}_{105} +1 t_x_{5}_{105} +1 t_x_{6}_{105} <= +1
- cf1_0: -1 t_x_{2}_{60} +1 t_x_{3}_{0} +1 t_x_{3}_{15} +1 t_x_{3}_{30} +1 t_x_{3}_{45} +1 t_x_{3}_{75}
- >= +0
- cf1_1: -1 t_x_{1}_{15} -1 t_x_{3}_{30} -1 t_x_{4}_{15} -1 t_x_{6}_{30} >= -3
-Bounds
- 0 <= t_x_{6}_{120} <= 0
- 0 <= t_x_{6}_{105} <= 0
- 0 <= t_x_{6}_{90} <= 0
- 0 <= t_x_{6}_{75} <= 1
- 0 <= t_x_{6}_{60} <= 1
- 0 <= t_x_{6}_{45} <= 1
- 0 <= t_x_{6}_{30} <= 1
- 0 <= t_x_{6}_{15} <= 1
- 0 <= t_x_{6}_{0} <= 1
- 0 <= t_x_{5}_{120} <= 0
- 0 <= t_x_{5}_{105} <= 0
- 0 <= t_x_{5}_{90} <= 0
- 0 <= t_x_{5}_{75} <= 1
- 0 <= t_x_{5}_{60} <= 1
- 0 <= t_x_{5}_{45} <= 1
- 0 <= t_x_{5}_{30} <= 1
- 0 <= t_x_{5}_{15} <= 1
- 0 <= t_x_{5}_{0} <= 1
- 0 <= t_x_{4}_{120} <= 0
- 0 <= t_x_{4}_{105} <= 0
- 0 <= t_x_{4}_{90} <= 0
- 0 <= t_x_{4}_{75} <= 1
- 0 <= t_x_{4}_{60} <= 1
- 0 <= t_x_{4}_{45} <= 1
- 0 <= t_x_{4}_{30} <= 1
- 0 <= t_x_{4}_{15} <= 1
- 0 <= t_x_{4}_{0} <= 1
- 0 <= t_x_{3}_{105} <= 0
- 0 <= t_x_{3}_{90} <= 0
- 0 <= t_x_{3}_{75} <= 1
- 0 <= t_x_{3}_{60} <= 1
- 0 <= t_x_{3}_{45} <= 1
- 0 <= t_x_{3}_{30} <= 1
- 0 <= t_x_{3}_{15} <= 1
- 0 <= t_x_{3}_{0} <= 1
- 0 <= t_x_{2}_{120} <= 0
- 0 <= t_x_{2}_{105} <= 0
- 0 <= t_x_{2}_{90} <= 0
- 0 <= t_x_{2}_{75} <= 1
- 0 <= t_x_{2}_{60} <= 1
- 0 <= t_x_{2}_{45} <= 1
- 0 <= t_x_{2}_{30} <= 1
- 0 <= t_x_{2}_{15} <= 1
- 0 <= t_x_{2}_{0} <= 1
- 0 <= t_x_{1}_{120} <= 0
- 0 <= t_x_{1}_{105} <= 0
- 0 <= t_x_{1}_{90} <= 0
- 0 <= t_x_{1}_{75} <= 0
- 0 <= t_x_{1}_{60} <= 1
- 0 <= t_x_{1}_{45} <= 1
- 0 <= t_x_{1}_{30} <= 1
- 0 <= t_x_{1}_{15} <= 1
- 0 <= t_x_{1}_{0} <= 1
- 0 <= t_x_{3}_{120} <= 0
-Binaries
- t_x_{6}_{120} t_x_{6}_{105} t_x_{6}_{90} t_x_{6}_{75} t_x_{6}_{60} t_x_{6}_{45} t_x_{6}_{30} t_x_{6}_{15}
- t_x_{6}_{0} t_x_{5}_{120} t_x_{5}_{105} t_x_{5}_{90} t_x_{5}_{75} t_x_{5}_{60} t_x_{5}_{45} t_x_{5}_{30}
- t_x_{5}_{15} t_x_{5}_{0} t_x_{4}_{120} t_x_{4}_{105} t_x_{4}_{90} t_x_{4}_{75} t_x_{4}_{60} t_x_{4}_{45}
- t_x_{4}_{30} t_x_{4}_{15} t_x_{4}_{0}
-End
+
+ row and column identified in (sec) : 0.000172
+
+ finished changing the variable's weight after (sec) : 0.000414
+
+ first iteration: problem solved after (sec) : 0.003492
+
+ subgradients found after (sec) : 0.008049
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.008091
+
+ finished changing the variable's weight after (sec) : 0.008106
+
+ first iteration: problem solved after (sec) : 0.010124
+
+ subgradients found after (sec) : 0.014033
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.014076
+
+ finished changing the variable's weight after (sec) : 0.014095
+
+ first iteration: problem solved after (sec) : 0.016643
+
+ subgradients found after (sec) : 0.026668
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.026862
+
+ finished changing the variable's weight after (sec) : 0.026919
+
+ first iteration: problem solved after (sec) : 0.033669
+
+ subgradients found after (sec) : 0.039234
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.039348
+
+ finished changing the variable's weight after (sec) : 0.039378
+
+ first iteration: problem solved after (sec) : 0.050431
+
+ subgradients found after (sec) : 0.063768
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.063828
+
+ finished changing the variable's weight after (sec) : 0.063845
+
+ first iteration: problem solved after (sec) : 0.069662
+
+ subgradients found after (sec) : 0.081571
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.081627
+
+ finished changing the variable's weight after (sec) : 0.081643
+
+ first iteration: problem solved after (sec) : 0.092082
+
+ subgradients found after (sec) : 0.098960
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.098998
+
+ finished changing the variable's weight after (sec) : 0.099017
+
+ first iteration: problem solved after (sec) : 0.110073
+
+ subgradients found after (sec) : 0.116468
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.116506
+
+ finished changing the variable's weight after (sec) : 0.116523
+
+ first iteration: problem solved after (sec) : 0.137771
+
+ subgradients found after (sec) : 0.146675
+, lowerbound = 46.000000
+
+ new dual found after (sec) : 0.146714
+
+ finished changing the variable's weight after (sec) : 0.146730
+
+ first iteration: problem solved after (sec) : 0.163891
+
+ subgradients found after (sec) : 0.170033
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.170076
+
+ finished changing the variable's weight after (sec) : 0.170090
+
+ first iteration: problem solved after (sec) : 0.185949
+
+ subgradients found after (sec) : 0.202449
+, lowerbound = 46.000000
+
+ new dual found after (sec) : 0.202493
+
+ finished changing the variable's weight after (sec) : 0.202560
+
+ first iteration: problem solved after (sec) : 0.220240
+
+ subgradients found after (sec) : 0.236488
+, lowerbound = 46.000000
+
+ new dual found after (sec) : 0.236539
+
+ finished changing the variable's weight after (sec) : 0.236588
+
+ first iteration: problem solved after (sec) : 0.247929
+
+ subgradients found after (sec) : 0.257980
+, lowerbound = 46.000000
+
+ new dual found after (sec) : 0.258025
+
+ finished changing the variable's weight after (sec) : 0.258043
+
+ first iteration: problem solved after (sec) : 0.267446
+
+ subgradients found after (sec) : 0.272147
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.272175
+
+ finished changing the variable's weight after (sec) : 0.272190
+
+ first iteration: problem solved after (sec) : 0.283181
+
+ subgradients found after (sec) : 0.292040
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.292135
+
+ finished changing the variable's weight after (sec) : 0.292170
+
+ first iteration: problem solved after (sec) : 0.304545
+
+ subgradients found after (sec) : 0.313874
+, lowerbound = 46.000000
+
+ new dual found after (sec) : 0.313925
+
+ finished changing the variable's weight after (sec) : 0.313937
+
+ first iteration: problem solved after (sec) : 0.325010
+
+ subgradients found after (sec) : 0.332073
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.332112
+
+ finished changing the variable's weight after (sec) : 0.332127
+
+ first iteration: problem solved after (sec) : 0.344750
+
+ subgradients found after (sec) : 0.350420
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.350457
+
+ finished changing the variable's weight after (sec) : 0.350477
+
+ first iteration: problem solved after (sec) : 0.362271
+
+ subgradients found after (sec) : 0.367667
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.367718
+
+ finished changing the variable's weight after (sec) : 0.367735
+
+ first iteration: problem solved after (sec) : 0.380156
+
+ subgradients found after (sec) : 0.389135
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.389198
+
+ finished changing the variable's weight after (sec) : 0.389223
+
+ first iteration: problem solved after (sec) : 0.403949
+
+ subgradients found after (sec) : 0.412378
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.412463
+
+ finished changing the variable's weight after (sec) : 0.412494
+
+ first iteration: problem solved after (sec) : 0.425490
+
+ subgradients found after (sec) : 0.436080
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.436199
+
+ finished changing the variable's weight after (sec) : 0.436247
+
+ first iteration: problem solved after (sec) : 0.448236
+
+ subgradients found after (sec) : 0.456060
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.456137
+
+ finished changing the variable's weight after (sec) : 0.456159
+
+ first iteration: problem solved after (sec) : 0.468406
+
+ subgradients found after (sec) : 0.480247
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.480289
+
+ finished changing the variable's weight after (sec) : 0.480304
+
+ first iteration: problem solved after (sec) : 0.490254
+
+ subgradients found after (sec) : 0.507246
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.507302
+
+ finished changing the variable's weight after (sec) : 0.507316
+
+ first iteration: problem solved after (sec) : 0.520940
+
+ subgradients found after (sec) : 0.530743
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.530785
+
+ finished changing the variable's weight after (sec) : 0.530803
+
+ first iteration: problem solved after (sec) : 0.544776
+
+ subgradients found after (sec) : 0.564726
+, lowerbound = 45.000000
+
+ new dual found after (sec) : 0.564781
+
+ finished changing the variable's weight after (sec) : 0.564805
+
+ first iteration: problem solved after (sec) : 0.580204
+
+ subgradients found after (sec) : 0.583687
+, lowerbound = 45.000000
+
+ new dual found after (sec) : 0.583727
+
+ finished changing the variable's weight after (sec) : 0.583743
+
+ first iteration: problem solved after (sec) : 0.592578
+
+ subgradients found after (sec) : 0.597884
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.597927
+
+ finished changing the variable's weight after (sec) : 0.597938
+
+ first iteration: problem solved after (sec) : 0.601858
+
+ subgradients found after (sec) : 0.613548
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.613673
+
+ finished changing the variable's weight after (sec) : 0.613776
+
+ first iteration: problem solved after (sec) : 0.618175
+
+ subgradients found after (sec) : 0.622886
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.623006
+
+ finished changing the variable's weight after (sec) : 0.623073
+
+ first iteration: problem solved after (sec) : 0.634738
+
+ subgradients found after (sec) : 0.644929
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.645087
+
+ finished changing the variable's weight after (sec) : 0.645164
+
+ first iteration: problem solved after (sec) : 0.650008
+
+ subgradients found after (sec) : 0.655097
+, lowerbound = 1.000000
+
+ new dual found after (sec) : 0.655218
+
+ finished changing the variable's weight after (sec) : 0.655246
+
+ first iteration: problem solved after (sec) : 0.663799
+
+ subgradients found after (sec) : 0.666612
+, lowerbound = 75.000000
+
+ new dual found after (sec) : 0.666659
+
+ finished changing the variable's weight after (sec) : 0.666675
+
+ first iteration: problem solved after (sec) : 0.673054
+
+ subgradients found after (sec) : 0.682264
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.682322
+
+ finished changing the variable's weight after (sec) : 0.682347
+
+ first iteration: problem solved after (sec) : 0.687952
+
+ subgradients found after (sec) : 0.697936
+, lowerbound = 45.000000
+
+ new dual found after (sec) : 0.697986
+
+ finished changing the variable's weight after (sec) : 0.698003
+
+ first iteration: problem solved after (sec) : 0.707602
+
+ subgradients found after (sec) : 0.714583
+, lowerbound = 45.000000
+
+ new dual found after (sec) : 0.714626
+
+ finished changing the variable's weight after (sec) : 0.714643
+
+ first iteration: problem solved after (sec) : 0.725393
+
+ subgradients found after (sec) : 0.730832
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.730879
+
+ finished changing the variable's weight after (sec) : 0.730895
+
+ first iteration: problem solved after (sec) : 0.746663
+
+ subgradients found after (sec) : 0.751705
+, lowerbound = 45.000000
+
+ new dual found after (sec) : 0.751794
+
+ finished changing the variable's weight after (sec) : 0.751817
+
+ first iteration: problem solved after (sec) : 0.763637
+
+ subgradients found after (sec) : 0.769296
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.769458
+
+ finished changing the variable's weight after (sec) : 0.769536
+
+ first iteration: problem solved after (sec) : 0.774887
+
+ subgradients found after (sec) : 0.783403
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.783515
+
+ finished changing the variable's weight after (sec) : 0.783556
+
+ first iteration: problem solved after (sec) : 0.789954
+
+ subgradients found after (sec) : 0.803887
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.803957
+
+ finished changing the variable's weight after (sec) : 0.803990
+
+ first iteration: problem solved after (sec) : 0.811035
+
+ subgradients found after (sec) : 0.820756
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.820823
+
+ finished changing the variable's weight after (sec) : 0.820861
+
+ first iteration: problem solved after (sec) : 0.831300
+
+ subgradients found after (sec) : 0.837327
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.837398
+
+ finished changing the variable's weight after (sec) : 0.837426
+
+ first iteration: problem solved after (sec) : 0.854492
+
+ subgradients found after (sec) : 0.864392
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.864545
+
+ finished changing the variable's weight after (sec) : 0.864614
+
+ first iteration: problem solved after (sec) : 0.881985
+
+ subgradients found after (sec) : 0.886445
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.886498
+
+ finished changing the variable's weight after (sec) : 0.886517
+
+ first iteration: problem solved after (sec) : 0.899798
+
+ subgradients found after (sec) : 0.903220
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.903264
+
+ finished changing the variable's weight after (sec) : 0.903286
+
+ first iteration: problem solved after (sec) : 0.914383
+
+ subgradients found after (sec) : 0.918506
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.918559
+
+ finished changing the variable's weight after (sec) : 0.918579
+
+ first iteration: problem solved after (sec) : 0.928877
+
+ subgradients found after (sec) : 0.939936
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.940008
+
+ finished changing the variable's weight after (sec) : 0.940026
+
+ first iteration: problem solved after (sec) : 0.947488
+
+ subgradients found after (sec) : 0.964042
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.964086
+
+ finished changing the variable's weight after (sec) : 0.964109
+
+ first iteration: problem solved after (sec) : 0.973338
+
+ subgradients found after (sec) : 0.980353
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 0.980432
+
+ finished changing the variable's weight after (sec) : 0.980456
+
+ first iteration: problem solved after (sec) : 0.987019
+
+ subgradients found after (sec) : 0.993122
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 0.993286
+
+ finished changing the variable's weight after (sec) : 0.993330
+
+ first iteration: problem solved after (sec) : 0.999261
+
+ subgradients found after (sec) : 1.005822
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.005879
+
+ finished changing the variable's weight after (sec) : 1.005898
+
+ first iteration: problem solved after (sec) : 1.012163
+
+ subgradients found after (sec) : 1.022449
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.022520
+
+ finished changing the variable's weight after (sec) : 1.022543
+
+ first iteration: problem solved after (sec) : 1.027397
+
+ subgradients found after (sec) : 1.039532
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.039577
+
+ finished changing the variable's weight after (sec) : 1.039593
+
+ first iteration: problem solved after (sec) : 1.046551
+
+ subgradients found after (sec) : 1.051828
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.051897
+
+ finished changing the variable's weight after (sec) : 1.051913
+
+ first iteration: problem solved after (sec) : 1.059183
+
+ subgradients found after (sec) : 1.062836
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.062881
+
+ finished changing the variable's weight after (sec) : 1.062899
+
+ first iteration: problem solved after (sec) : 1.077821
+
+ subgradients found after (sec) : 1.081575
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.081611
+
+ finished changing the variable's weight after (sec) : 1.081627
+
+ first iteration: problem solved after (sec) : 1.096263
+
+ subgradients found after (sec) : 1.099700
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.099747
+
+ finished changing the variable's weight after (sec) : 1.099770
+
+ first iteration: problem solved after (sec) : 1.115608
+
+ subgradients found after (sec) : 1.119333
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.119369
+
+ finished changing the variable's weight after (sec) : 1.119384
+
+ first iteration: problem solved after (sec) : 1.134438
+
+ subgradients found after (sec) : 1.138069
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.138116
+
+ finished changing the variable's weight after (sec) : 1.138133
+
+ first iteration: problem solved after (sec) : 1.158348
+
+ subgradients found after (sec) : 1.166183
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.166226
+
+ finished changing the variable's weight after (sec) : 1.166244
+
+ first iteration: problem solved after (sec) : 1.181479
+
+ subgradients found after (sec) : 1.184905
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.184957
+
+ finished changing the variable's weight after (sec) : 1.184983
+
+ first iteration: problem solved after (sec) : 1.200687
+
+ subgradients found after (sec) : 1.207300
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.207356
+
+ finished changing the variable's weight after (sec) : 1.207385
+
+ first iteration: problem solved after (sec) : 1.219406
+
+ subgradients found after (sec) : 1.222643
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.222679
+
+ finished changing the variable's weight after (sec) : 1.222696
+
+ first iteration: problem solved after (sec) : 1.238111
+
+ subgradients found after (sec) : 1.244475
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.244587
+
+ finished changing the variable's weight after (sec) : 1.244842
+
+ first iteration: problem solved after (sec) : 1.257716
+
+ subgradients found after (sec) : 1.273762
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.273810
+
+ finished changing the variable's weight after (sec) : 1.273827
+
+ first iteration: problem solved after (sec) : 1.285971
+
+ subgradients found after (sec) : 1.290857
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.290905
+
+ finished changing the variable's weight after (sec) : 1.290927
+
+ first iteration: problem solved after (sec) : 1.311756
+
+ subgradients found after (sec) : 1.326074
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.326125
+
+ finished changing the variable's weight after (sec) : 1.326160
+
+ first iteration: problem solved after (sec) : 1.334530
+
+ subgradients found after (sec) : 1.349052
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.349107
+
+ finished changing the variable's weight after (sec) : 1.349124
+
+ first iteration: problem solved after (sec) : 1.353694
+
+ subgradients found after (sec) : 1.364390
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.364442
+
+ finished changing the variable's weight after (sec) : 1.364465
+
+ first iteration: problem solved after (sec) : 1.368809
+
+ subgradients found after (sec) : 1.381127
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.381194
+
+ finished changing the variable's weight after (sec) : 1.381211
+
+ first iteration: problem solved after (sec) : 1.390071
+
+ subgradients found after (sec) : 1.398892
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.398947
+
+ finished changing the variable's weight after (sec) : 1.399007
+
+ first iteration: problem solved after (sec) : 1.412047
+
+ subgradients found after (sec) : 1.417008
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.417149
+
+ finished changing the variable's weight after (sec) : 1.417165
+
+ first iteration: problem solved after (sec) : 1.429299
+
+ subgradients found after (sec) : 1.432734
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.432779
+
+ finished changing the variable's weight after (sec) : 1.432807
+
+ first iteration: problem solved after (sec) : 1.439484
+
+ subgradients found after (sec) : 1.452770
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.452826
+
+ finished changing the variable's weight after (sec) : 1.452858
+
+ first iteration: problem solved after (sec) : 1.466494
+
+ subgradients found after (sec) : 1.470024
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.470077
+
+ finished changing the variable's weight after (sec) : 1.470098
+
+ first iteration: problem solved after (sec) : 1.482338
+
+ subgradients found after (sec) : 1.488006
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.488054
+
+ finished changing the variable's weight after (sec) : 1.488087
+
+ first iteration: problem solved after (sec) : 1.504377
+
+ subgradients found after (sec) : 1.507797
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.507849
+
+ finished changing the variable's weight after (sec) : 1.507864
+
+ first iteration: problem solved after (sec) : 1.512105
+
+ subgradients found after (sec) : 1.520699
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.520751
+
+ finished changing the variable's weight after (sec) : 1.520767
+
+ first iteration: problem solved after (sec) : 1.528500
+
+ subgradients found after (sec) : 1.541864
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.542110
+
+ finished changing the variable's weight after (sec) : 1.542185
+
+ first iteration: problem solved after (sec) : 1.552735
+
+ subgradients found after (sec) : 1.564996
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.565076
+
+ finished changing the variable's weight after (sec) : 1.565098
+
+ first iteration: problem solved after (sec) : 1.580799
+
+ subgradients found after (sec) : 1.585080
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.585129
+
+ finished changing the variable's weight after (sec) : 1.585152
+
+ first iteration: problem solved after (sec) : 1.599165
+
+ subgradients found after (sec) : 1.602631
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.602675
+
+ finished changing the variable's weight after (sec) : 1.602692
+
+ first iteration: problem solved after (sec) : 1.616960
+
+ subgradients found after (sec) : 1.621737
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.621923
+
+ finished changing the variable's weight after (sec) : 1.621995
+
+ first iteration: problem solved after (sec) : 1.636517
+
+ subgradients found after (sec) : 1.641711
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.641970
+
+ finished changing the variable's weight after (sec) : 1.642051
+
+ first iteration: problem solved after (sec) : 1.656004
+
+ subgradients found after (sec) : 1.659406
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.659453
+
+ finished changing the variable's weight after (sec) : 1.659469
+
+ first iteration: problem solved after (sec) : 1.675267
+
+ subgradients found after (sec) : 1.678431
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.678479
+
+ finished changing the variable's weight after (sec) : 1.678497
+
+ first iteration: problem solved after (sec) : 1.685132
+
+ subgradients found after (sec) : 1.696662
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.696718
+
+ finished changing the variable's weight after (sec) : 1.696772
+
+ first iteration: problem solved after (sec) : 1.707206
+
+ subgradients found after (sec) : 1.715269
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.715314
+
+ finished changing the variable's weight after (sec) : 1.715328
+
+ first iteration: problem solved after (sec) : 1.730080
+
+ subgradients found after (sec) : 1.734707
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.734849
+
+ finished changing the variable's weight after (sec) : 1.734881
+
+ first iteration: problem solved after (sec) : 1.744065
+
+ subgradients found after (sec) : 1.757242
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.757308
+
+ finished changing the variable's weight after (sec) : 1.757331
+
+ first iteration: problem solved after (sec) : 1.772778
+
+ subgradients found after (sec) : 1.776581
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.776661
+
+ finished changing the variable's weight after (sec) : 1.776715
+
+ first iteration: problem solved after (sec) : 1.788041
+
+ subgradients found after (sec) : 1.791603
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.791644
+
+ finished changing the variable's weight after (sec) : 1.791666
+
+ first iteration: problem solved after (sec) : 1.804485
+
+ subgradients found after (sec) : 1.811822
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.811865
+
+ finished changing the variable's weight after (sec) : 1.811887
+
+ first iteration: problem solved after (sec) : 1.816012
+
+ subgradients found after (sec) : 1.831486
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.831672
+
+ finished changing the variable's weight after (sec) : 1.831734
+
+ first iteration: problem solved after (sec) : 1.837399
+
+ subgradients found after (sec) : 1.852032
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.852074
+
+ finished changing the variable's weight after (sec) : 1.852090
+
+ first iteration: problem solved after (sec) : 1.858960
+
+ subgradients found after (sec) : 1.874654
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.874708
+
+ finished changing the variable's weight after (sec) : 1.874723
+
+ first iteration: problem solved after (sec) : 1.881609
+
+ subgradients found after (sec) : 1.894558
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.894651
+
+ finished changing the variable's weight after (sec) : 1.894701
+
+ first iteration: problem solved after (sec) : 1.914189
+
+ subgradients found after (sec) : 1.918536
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.918665
+
+ finished changing the variable's weight after (sec) : 1.918728
+
+ first iteration: problem solved after (sec) : 1.932328
+
+ subgradients found after (sec) : 1.942495
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.942625
+
+ finished changing the variable's weight after (sec) : 1.942659
+
+ first iteration: problem solved after (sec) : 1.954691
+
+ subgradients found after (sec) : 1.969142
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 1.969201
+
+ finished changing the variable's weight after (sec) : 1.969223
+
+ first iteration: problem solved after (sec) : 1.975786
+
+ subgradients found after (sec) : 1.993815
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 1.993863
+
+ finished changing the variable's weight after (sec) : 1.993878
+
+ first iteration: problem solved after (sec) : 2.008093
+
+ subgradients found after (sec) : 2.019121
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.019164
+
+ finished changing the variable's weight after (sec) : 2.019180
+
+ first iteration: problem solved after (sec) : 2.030906
+
+ subgradients found after (sec) : 2.034102
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.034144
+
+ finished changing the variable's weight after (sec) : 2.034160
+
+ first iteration: problem solved after (sec) : 2.045757
+
+ subgradients found after (sec) : 2.059330
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.059665
+
+ finished changing the variable's weight after (sec) : 2.059770
+
+ first iteration: problem solved after (sec) : 2.067335
+
+ subgradients found after (sec) : 2.076437
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.076619
+
+ finished changing the variable's weight after (sec) : 2.076666
+
+ first iteration: problem solved after (sec) : 2.084611
+
+ subgradients found after (sec) : 2.092310
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.092434
+
+ finished changing the variable's weight after (sec) : 2.092461
+
+ first iteration: problem solved after (sec) : 2.106273
+
+ subgradients found after (sec) : 2.113490
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.114016
+
+ finished changing the variable's weight after (sec) : 2.114102
+
+ first iteration: problem solved after (sec) : 2.137330
+
+ subgradients found after (sec) : 2.140904
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.140941
+
+ finished changing the variable's weight after (sec) : 2.140958
+
+ first iteration: problem solved after (sec) : 2.155801
+
+ subgradients found after (sec) : 2.160276
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.160327
+
+ finished changing the variable's weight after (sec) : 2.160345
+
+ first iteration: problem solved after (sec) : 2.173429
+
+ subgradients found after (sec) : 2.176844
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.176889
+
+ finished changing the variable's weight after (sec) : 2.176905
+
+ first iteration: problem solved after (sec) : 2.189498
+
+ subgradients found after (sec) : 2.193006
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.193080
+
+ finished changing the variable's weight after (sec) : 2.193122
+
+ first iteration: problem solved after (sec) : 2.206445
+
+ subgradients found after (sec) : 2.216242
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.216291
+
+ finished changing the variable's weight after (sec) : 2.216308
+
+ first iteration: problem solved after (sec) : 2.227342
+
+ subgradients found after (sec) : 2.241271
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.241314
+
+ finished changing the variable's weight after (sec) : 2.241332
+
+ first iteration: problem solved after (sec) : 2.246007
+
+ subgradients found after (sec) : 2.249583
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.249640
+
+ finished changing the variable's weight after (sec) : 2.249656
+
+ first iteration: problem solved after (sec) : 2.265344
+
+ subgradients found after (sec) : 2.269057
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.269077
+
+ finished changing the variable's weight after (sec) : 2.269093
+
+ first iteration: problem solved after (sec) : 2.297654
+
+ subgradients found after (sec) : 2.303096
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.303371
+
+ finished changing the variable's weight after (sec) : 2.303458
+
+ first iteration: problem solved after (sec) : 2.318736
+
+ subgradients found after (sec) : 2.322329
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.322373
+
+ finished changing the variable's weight after (sec) : 2.322388
+
+ first iteration: problem solved after (sec) : 2.340886
+
+ subgradients found after (sec) : 2.358288
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.358459
+
+ finished changing the variable's weight after (sec) : 2.358501
+
+ first iteration: problem solved after (sec) : 2.365414
+
+ subgradients found after (sec) : 2.392701
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.392796
+
+ finished changing the variable's weight after (sec) : 2.392824
+
+ first iteration: problem solved after (sec) : 2.397168
+
+ subgradients found after (sec) : 2.400464
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.400518
+
+ finished changing the variable's weight after (sec) : 2.400535
+
+ first iteration: problem solved after (sec) : 2.416346
+
+ subgradients found after (sec) : 2.422995
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.423040
+
+ finished changing the variable's weight after (sec) : 2.423055
+
+ first iteration: problem solved after (sec) : 2.440188
+
+ subgradients found after (sec) : 2.443503
+, lowerbound = 120.000000
+
+ new dual found after (sec) : 2.443540
+
+ finished changing the variable's weight after (sec) : 2.443556
+
+ first iteration: problem solved after (sec) : 2.458354
+
+ subgradients found after (sec) : 2.464748
+, lowerbound = 31.000000
+
+ new dual found after (sec) : 2.464921
diff --git a/var.txt b/var.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/varobjs.txt b/varobjs.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391