Reading training examples...done
Training set properties: 23 features, 90 rankings, 8184 examples
NOTE: Adjusted stopping criterion relative to maximum loss: eps=0.486067
Iter 1: .........*(NumConst=1, SV=1, CEps=486.0667, QPEps=0.0000)
Iter 2: .........*(NumConst=2, SV=2, CEps=430.5748, QPEps=0.0001)
Iter 3: .........*(NumConst=3, SV=3, CEps=790.5531, QPEps=0.0008)
Iter 4: .........*(NumConst=4, SV=3, CEps=2043.8807, QPEps=0.0001)
Iter 5: .........*(NumConst=5, SV=5, CEps=1143.9372, QPEps=0.0012)
Iter 6: .........*(NumConst=6, SV=6, CEps=146.7187, QPEps=0.0021)
Iter 7: .........*(NumConst=7, SV=7, CEps=130.5926, QPEps=43.9826)
Iter 8: .........*(NumConst=8, SV=8, CEps=80.6832, QPEps=40.2432)
Iter 9: .........*(NumConst=9, SV=7, CEps=141.3987, QPEps=39.6042)
Iter 10: .........*(NumConst=10, SV=6, CEps=58.4636, QPEps=27.1044)
Iter 11: .........*(NumConst=11, SV=7, CEps=94.4660, QPEps=28.1428)
Iter 12: .........*(NumConst=12, SV=7, CEps=53.6284, QPEps=0.0000)
Iter 13: .........*(NumConst=13, SV=7, CEps=29.4791, QPEps=0.0000)
Iter 14: .........*(NumConst=14, SV=7, CEps=25.4413, QPEps=0.0000)
Iter 15: .........*(NumConst=15, SV=8, CEps=26.6021, QPEps=11.9274)
Iter 16: .........*(NumConst=16, SV=8, CEps=20.8265, QPEps=9.3152)
Iter 17: .........*(NumConst=17, SV=8, CEps=12.8114, QPEps=3.2761)
Iter 18: .........*(NumConst=18, SV=8, CEps=12.0119, QPEps=5.4897)
Iter 19: .........*(NumConst=19, SV=8, CEps=10.9110, QPEps=2.3233)
Iter 20: .........*(NumConst=20, SV=7, CEps=6.7959, QPEps=3.0261)
Iter 21: .........*(NumConst=21, SV=8, CEps=6.5674, QPEps=2.4143)
Iter 22: .........*(NumConst=22, SV=8, CEps=7.7027, QPEps=3.2482)
Iter 23: .........*(NumConst=23, SV=7, CEps=5.5158, QPEps=0.9177)
Iter 24: .........*(NumConst=24, SV=7, CEps=2.8470, QPEps=1.3323)
Iter 25: .........*(NumConst=25, SV=7, CEps=4.8105, QPEps=0.5275)
Iter 26: .........*(NumConst=26, SV=7, CEps=4.3174, QPEps=0.0527)
Iter 27: .........*(NumConst=27, SV=7, CEps=1.7090, QPEps=0.5724)
Iter 28: .........*(NumConst=28, SV=8, CEps=2.4975, QPEps=0.1659)
Iter 29: .........*(NumConst=29, SV=7, CEps=2.7247, QPEps=0.0000)
Iter 30: .........*(NumConst=30, SV=7, CEps=2.1076, QPEps=0.0000)
Iter 31: .........*(NumConst=31, SV=7, CEps=1.0915, QPEps=0.0000)
Iter 32: .........*(NumConst=32, SV=9, CEps=1.0475, QPEps=0.5144)
Iter 33: .........*(NumConst=33, SV=8, CEps=1.7466, QPEps=0.2101)
Iter 34: .........*(NumConst=34, SV=7, CEps=1.2175, QPEps=0.0000)
Iter 35: .........*(NumConst=35, SV=7, CEps=0.6166, QPEps=0.1387)
Iter 36: .........*(NumConst=36, SV=7, CEps=0.8213, QPEps=0.2349)
Iter 37: .........(NumConst=36, SV=7, CEps=0.4095, QPEps=0.2349)
Final epsilon on KKT-Conditions: 0.40952
Upper bound on duality gap: 0.01373
Dual objective value: dval=8.93416
Primal objective value: pval=8.94789
Total number of constraints in final working set: 36 (of 36)
Number of iterations: 37
Number of calls to 'find_most_violated_constraint': 3330
Number of SV: 7 
Norm of weight vector: |w|=0.99735
Value of slack variable (on working set): xi=281.27574
Value of slack variable (global): xi=281.68453
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=3385.21254
Runtime in cpu-seconds: 0.22
Compacting linear model...done
Writing learned model...done