Reading training examples...done
Training set properties: 23 features, 90 rankings, 8132 examples
NOTE: Adjusted stopping criterion relative to maximum loss: eps=0.528367
Iter 1: .........*(NumConst=1, SV=1, CEps=528.3667, QPEps=0.0000)
Iter 2: .........*(NumConst=2, SV=2, CEps=492.9135, QPEps=0.0003)
Iter 3: .........*(NumConst=3, SV=3, CEps=922.4975, QPEps=0.0015)
Iter 4: .........*(NumConst=4, SV=3, CEps=2813.1056, QPEps=0.0045)
Iter 5: .........*(NumConst=5, SV=4, CEps=989.9420, QPEps=0.0007)
Iter 6: .........*(NumConst=6, SV=5, CEps=486.1549, QPEps=46.0564)
Iter 7: .........*(NumConst=7, SV=5, CEps=135.5902, QPEps=48.8195)
Iter 8: .........*(NumConst=8, SV=6, CEps=189.4763, QPEps=49.5231)
Iter 9: .........*(NumConst=9, SV=6, CEps=101.0990, QPEps=11.9544)
Iter 10: .........*(NumConst=10, SV=6, CEps=66.9774, QPEps=0.0002)
Iter 11: .........*(NumConst=11, SV=6, CEps=61.0998, QPEps=2.0429)
Iter 12: .........*(NumConst=12, SV=5, CEps=63.9973, QPEps=0.0010)
Iter 13: .........*(NumConst=13, SV=6, CEps=25.6032, QPEps=0.0001)
Iter 14: .........*(NumConst=14, SV=7, CEps=72.5986, QPEps=4.0423)
Iter 15: .........*(NumConst=15, SV=7, CEps=30.5543, QPEps=1.1314)
Iter 16: .........*(NumConst=16, SV=6, CEps=25.7225, QPEps=10.9938)
Iter 17: .........*(NumConst=17, SV=6, CEps=12.8336, QPEps=0.0001)
Iter 18: .........*(NumConst=18, SV=6, CEps=14.1243, QPEps=0.0000)
Iter 19: .........*(NumConst=19, SV=6, CEps=12.1035, QPEps=5.7905)
Iter 20: .........*(NumConst=20, SV=6, CEps=12.6845, QPEps=0.0000)
Iter 21: .........*(NumConst=21, SV=6, CEps=7.2979, QPEps=1.4697)
Iter 22: .........*(NumConst=22, SV=6, CEps=7.6480, QPEps=0.0000)
Iter 23: .........*(NumConst=23, SV=7, CEps=5.0039, QPEps=1.8588)
Iter 24: .........*(NumConst=24, SV=7, CEps=4.7318, QPEps=0.0000)
Iter 25: .........*(NumConst=25, SV=6, CEps=4.1982, QPEps=1.8511)
Iter 26: .........*(NumConst=26, SV=6, CEps=3.3821, QPEps=0.0000)
Iter 27: .........*(NumConst=27, SV=6, CEps=1.9688, QPEps=0.0000)
Iter 28: .........*(NumConst=28, SV=7, CEps=1.9548, QPEps=0.7913)
Iter 29: .........*(NumConst=29, SV=6, CEps=3.4926, QPEps=0.4929)
Iter 30: .........*(NumConst=30, SV=6, CEps=1.4217, QPEps=0.0000)
Iter 31: .........*(NumConst=31, SV=5, CEps=1.1743, QPEps=0.5660)
Iter 32: .........*(NumConst=32, SV=5, CEps=0.9450, QPEps=0.0000)
Iter 33: .........*(NumConst=33, SV=6, CEps=0.5624, QPEps=0.0988)
Iter 34: .........*(NumConst=34, SV=6, CEps=1.0349, QPEps=0.1383)
Iter 35: .........*(NumConst=35, SV=6, CEps=0.8099, QPEps=0.0000)
Iter 36: .........*(NumConst=36, SV=6, CEps=0.5404, QPEps=0.1769)
Iter 37: .........(NumConst=36, SV=6, CEps=0.4295, QPEps=0.1769)
Final epsilon on KKT-Conditions: 0.42949
Upper bound on duality gap: 0.01332
Dual objective value: dval=9.92378
Primal objective value: pval=9.93710
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: 6 
Norm of weight vector: |w|=1.04929
Value of slack variable (on working set): xi=312.45706
Value of slack variable (global): xi=312.88652
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=3492.85876
Runtime in cpu-seconds: 0.20
Compacting linear model...done
Writing learned model...done