Reading training examples...done
Training set properties: 19 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=388.2542, QPEps=0.0000)
Iter 3: .........*(NumConst=3, SV=3, CEps=95.2484, QPEps=20.9085)
Iter 4: .........*(NumConst=4, SV=3, CEps=36.1785, QPEps=14.4051)
Iter 5: .........*(NumConst=5, SV=3, CEps=27.7502, QPEps=12.1233)
Iter 6: .........*(NumConst=6, SV=3, CEps=9.4345, QPEps=4.0354)
Iter 7: .........*(NumConst=7, SV=3, CEps=2.1859, QPEps=0.4106)
Iter 8: .........*(NumConst=8, SV=2, CEps=1.6771, QPEps=0.5924)
Iter 9: .........*(NumConst=9, SV=3, CEps=0.8015, QPEps=0.2864)
Iter 10: .........(NumConst=9, SV=3, CEps=0.4310, QPEps=0.2864)
Final epsilon on KKT-Conditions: 0.43099
Upper bound on duality gap: 0.00056
Dual objective value: dval=0.33355
Primal objective value: pval=0.33411
Total number of constraints in final working set: 9 (of 9)
Number of iterations: 10
Number of calls to 'find_most_violated_constraint': 900
Number of SV: 3 
Norm of weight vector: |w|=0.14651
Value of slack variable (on working set): xi=322.94683
Value of slack variable (global): xi=323.37782
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=2671.05679
Runtime in cpu-seconds: 0.02
Compacting linear model...done
Writing learned model...done