MLcomp version alpha Statistics: 1208 users, 401 datasets, 328 programs, 9937 runs (0 queued, 0 running), 0 workers

Run 1805

Creator	pliang
Program	svmlight_multiclass-linear
Dataset	isolet
Task type	MulticlassClassification
Created	1y122d ago
Download	Login required!

Status: Done!

Total Time	1m27s
Max Memory Usage	265M
Domain:	MulticlassClassification
Learn time
Train error	0.108
Predict train time
Test error	0.131
Predict test time

Log file

... (lines omitted) ...
Iter 30: .........*(NumConst=30, SV=15, CEps=0.1038, QPEps=0.0373)
Iter 31: *(NumConst=31, SV=15, CEps=0.1313, QPEps=0.0413)
Iter 32: *(NumConst=32, SV=14, CEps=0.1033, QPEps=0.0242)
Iter 33: *(NumConst=33, SV=15, CEps=0.1229, QPEps=0.0412)
Iter 34: .........(NumConst=33, SV=15, CEps=0.0924, QPEps=0.0412)
Final epsilon on KKT-Conditions: 0.09244
Upper bound on duality gap: 0.09027
Dual objective value: dval=99.81251
Primal objective value: pval=99.90278
Total number of constraints in final working set: 33 (of 33)
Number of iterations: 34
Number of calls to 'find_most_violated_constraint': 43670
Number of SV: 15 
Norm of weight vector: |w|=0.61235
Value of slack variable (on working set): xi=99.64972
Value of slack variable (global): xi=99.71530
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=10.73772
Runtime in cpu-seconds: 6.68
Final number of constraints in cache: 21833
Compacting linear model...done
Writing learned model...done
=== END _tune-hyperparameter2: ./run learn ../cv.train --- OK [10s]
=== START _tune-hyperparameter2: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions2
Reading model...done.
Reading test examples... (1871 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.13
Average loss on test set: 29.8236
Zero/one-error on test set: 29.82% (1313 correct, 558 incorrect, 1871 total)
=== END _tune-hyperparameter2: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions2 --- OK [2s]
=== START program5: ./run evaluate ../program1/cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions2
=== END program5: ./run evaluate ../program1/cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions2 --- OK [1s]
CV error rate 0.298236237306253 with hyperparameter 1.0

===== Cross-validator: trying hyperparameter 10.0 =====
=== START _tune-hyperparameter3: ./run setHyperparameter 10.0
=== END _tune-hyperparameter3: ./run setHyperparameter 10.0 --- OK [0s]
=== START _tune-hyperparameter3: ./run learn ../cv.train
Using hyperparameter c = 10.0
Reading training examples... (4367 examples) done
Training set properties: 617 features, 26 classes
Iter 1: .........*(NumConst=1, SV=1, CEps=100.0000, QPEps=0.0000)
Iter 2: .........*(NumConst=2, SV=2, CEps=104.7955, QPEps=0.0000)
Iter 3: *(NumConst=3, SV=3, CEps=48.0271, QPEps=0.0000)
Iter 4: .........*(NumConst=4, SV=4, CEps=91.3925, QPEps=0.0000)
Iter 5: *(NumConst=5, SV=4, CEps=30.7971, QPEps=0.0000)
Iter 6: *(NumConst=6, SV=4, CEps=10.9269, QPEps=0.0000)
Iter 7: *(NumConst=7, SV=4, CEps=9.3750, QPEps=0.0000)
Iter 8: .........*(NumConst=8, SV=6, CEps=50.9990, QPEps=2.2849)
Iter 9: *(NumConst=9, SV=5, CEps=17.8556, QPEps=0.0002)
Iter 10: *(NumConst=10, SV=5, CEps=19.1944, QPEps=0.0000)
Iter 11: *(NumConst=11, SV=5, CEps=10.0858, QPEps=0.0000)
Iter 12: *(NumConst=12, SV=5, CEps=13.2373, QPEps=0.0000)
Iter 13: *(NumConst=13, SV=6, CEps=5.8743, QPEps=0.0000)
Iter 14: .........*(NumConst=14, SV=8, CEps=34.8326, QPEps=9.4531)
Iter 15: *(NumConst=15, SV=7, CEps=9.3952, QPEps=0.0000)
Iter 16: *(NumConst=16, SV=9, CEps=4.6937, QPEps=2.1008)
Iter 17: *(NumConst=17, SV=7, CEps=4.2188, QPEps=0.0000)
Iter 18: *(NumConst=18, SV=8, CEps=3.7798, QPEps=1.2456)
Iter 19: .........*(NumConst=19, SV=10, CEps=15.5813, QPEps=1.3866)
Iter 20: *(NumConst=20, SV=10, CEps=4.3855, QPEps=1.9623)
Iter 21: *(NumConst=21, SV=9, CEps=4.6114, QPEps=1.3090)
Iter 22: *(NumConst=22, SV=9, CEps=2.2485, QPEps=0.1961)
Iter 23: *(NumConst=23, SV=11, CEps=1.9011, QPEps=0.6641)
Iter 24: *(NumConst=24, SV=9, CEps=2.2987, QPEps=0.8671)
Iter 25: .........*(NumConst=25, SV=10, CEps=5.5770, QPEps=0.0000)
Iter 26: *(NumConst=26, SV=10, CEps=2.1040, QPEps=0.6633)
Iter 27: *(NumConst=27, SV=11, CEps=1.6677, QPEps=0.1619)
Iter 28: *(NumConst=28, SV=13, CEps=1.3328, QPEps=0.2770)
Iter 29: *(NumConst=29, SV=12, CEps=1.6419, QPEps=0.6162)
Iter 30: *(NumConst=30, SV=13, CEps=0.9190, QPEps=0.3144)
Iter 31: *(NumConst=31, SV=11, CEps=1.2627, QPEps=0.4546)
Iter 32: *(NumConst=32, SV=12, CEps=0.9893, QPEps=0.4166)
Iter 33: *(NumConst=33, SV=11, CEps=0.8936, QPEps=0.0472)
Iter 34: *(NumConst=34, SV=11, CEps=0.6562, QPEps=0.0520)
Iter 35: *(NumConst=35, SV=11, CEps=0.6794, QPEps=0.2240)
Iter 36: *(NumConst=36, SV=13, CEps=0.7552, QPEps=0.2226)
Iter 37: *(NumConst=37, SV=12, CEps=0.7110, QPEps=0.1285)
Iter 38: .........*(NumConst=38, SV=13, CEps=1.6718, QPEps=0.1736)
Iter 39: *(NumConst=39, SV=14, CEps=0.9192, QPEps=0.3560)
Iter 40: *(NumConst=40, SV=14, CEps=0.8901, QPEps=0.3262)
Iter 41: *(NumConst=41, SV=14, CEps=0.6758, QPEps=0.2089)
Iter 42: *(NumConst=42, SV=13, CEps=0.5572, QPEps=0.2042)
Iter 43: *(NumConst=43, SV=15, CEps=0.6592, QPEps=0.0995)
Iter 44: *(NumConst=44, SV=15, CEps=0.5364, QPEps=0.2205)
Iter 45: *(NumConst=45, SV=15, CEps=0.6548, QPEps=0.1613)
Iter 46: *(NumConst=46, SV=16, CEps=0.4996, QPEps=0.2447)
Iter 47: *(NumConst=47, SV=15, CEps=0.6162, QPEps=0.1571)
Iter 48: *(NumConst=48, SV=17, CEps=0.4060, QPEps=0.1770)
Iter 49: *(NumConst=49, SV=15, CEps=0.4357, QPEps=0.1348)
Iter 50: *(NumConst=50, SV=16, CEps=0.4667, QPEps=0.1731)
Iter 51: *(NumConst=51, SV=14, CEps=0.3557, QPEps=0.0932)
Iter 52: *(NumConst=52, SV=14, CEps=0.4173, QPEps=0.1378)
Iter 53: *(NumConst=53, SV=14, CEps=0.2549, QPEps=0.0778)
Iter 54: *(NumConst=53, SV=15, CEps=0.4919, QPEps=0.0739)
Iter 55: *(NumConst=54, SV=15, CEps=0.2621, QPEps=0.0656)
Iter 56: *(NumConst=54, SV=14, CEps=0.2877, QPEps=0.0849)
Iter 57: *(NumConst=55, SV=15, CEps=0.2320, QPEps=0.0909)
Iter 58: *(NumConst=55, SV=14, CEps=0.2659, QPEps=0.0776)
Iter 59: *(NumConst=56, SV=14, CEps=0.1709, QPEps=0.0405)
Iter 60: *(NumConst=57, SV=16, CEps=0.2747, QPEps=0.0753)
Iter 61: *(NumConst=58, SV=16, CEps=0.2613, QPEps=0.0850)
Iter 62: *(NumConst=59, SV=15, CEps=0.2076, QPEps=0.0695)
Iter 63: *(NumConst=60, SV=15, CEps=0.1808, QPEps=0.0595)
Iter 64: *(NumConst=59, SV=16, CEps=0.1985, QPEps=0.0707)
Iter 65: *(NumConst=60, SV=15, CEps=0.1798, QPEps=0.0478)
Iter 66: .........*(NumConst=60, SV=16, CEps=0.2778, QPEps=0.0900)
Iter 67: *(NumConst=61, SV=18, CEps=0.2219, QPEps=0.0879)
Iter 68: *(NumConst=62, SV=17, CEps=0.2984, QPEps=0.0799)
Iter 69: *(NumConst=63, SV=18, CEps=0.1888, QPEps=0.0759)
Iter 70: *(NumConst=62, SV=19, CEps=0.2255, QPEps=0.0733)
Iter 71: *(NumConst=63, SV=19, CEps=0.3502, QPEps=0.0899)
Iter 72: *(NumConst=64, SV=20, CEps=0.1761, QPEps=0.0646)
Iter 73: *(NumConst=62, SV=21, CEps=0.1585, QPEps=0.0470)
Iter 74: *(NumConst=63, SV=16, CEps=0.2579, QPEps=0.0723)
Iter 75: *(NumConst=64, SV=16, CEps=0.1604, QPEps=0.0670)
Iter 76: *(NumConst=64, SV=16, CEps=0.1353, QPEps=0.0312)
Iter 77: *(NumConst=64, SV=16, CEps=0.1824, QPEps=0.0321)
Iter 78: *(NumConst=63, SV=16, CEps=0.1492, QPEps=0.0593)
Iter 79: .........*(NumConst=63, SV=17, CEps=0.1206, QPEps=0.0495)
Iter 80: *(NumConst=61, SV=16, CEps=0.1721, QPEps=0.0346)
Iter 81: *(NumConst=62, SV=17, CEps=0.1212, QPEps=0.0336)
Iter 82: *(NumConst=61, SV=18, CEps=0.1807, QPEps=0.0464)
Iter 83: *(NumConst=62, SV=19, CEps=0.1518, QPEps=0.0546)
Iter 84: *(NumConst=62, SV=19, CEps=0.1105, QPEps=0.0390)
Iter 85: .........(NumConst=62, SV=19, CEps=0.0883, QPEps=0.0390)
Final epsilon on KKT-Conditions: 0.08831
Upper bound on duality gap: 0.89233
Dual objective value: dval=984.39623
Primal objective value: pval=985.28856
Total number of constraints in final working set: 62 (of 84)
Number of iterations: 85
Number of calls to 'find_most_violated_constraint': 48037
Number of SV: 19 
Norm of weight vector: |w|=5.58637
Value of slack variable (on working set): xi=96.91271
Value of slack variable (global): xi=96.96848
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=10.73772
Runtime in cpu-seconds: 12.72
Final number of constraints in cache: 21835
Compacting linear model...done
Writing learned model...done
=== END _tune-hyperparameter3: ./run learn ../cv.train --- OK [16s]
=== START _tune-hyperparameter3: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions3
Reading model...done.
Reading test examples... (1871 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.12
Average loss on test set: 14.0567
Zero/one-error on test set: 14.06% (1608 correct, 263 incorrect, 1871 total)
=== END _tune-hyperparameter3: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions3 --- OK [2s]
=== START program5: ./run evaluate ../program1/cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions3
=== END program5: ./run evaluate ../program1/cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions3 --- OK [0s]
CV error rate 0.140566541956173 with hyperparameter 10.0

===== Cross-validator: trying hyperparameter 100.0 =====
=== START _tune-hyperparameter4: ./run setHyperparameter 100.0
=== END _tune-hyperparameter4: ./run setHyperparameter 100.0 --- OK [0s]
=== START _tune-hyperparameter4: ./run learn ../cv.train
Using hyperparameter c = 100.0
Reading training examples... (4367 examples) done
Training set properties: 617 features, 26 classes
Iter 1: .........*(NumConst=1, SV=1, CEps=100.0000, QPEps=0.0000)
Iter 2: .........*(NumConst=2, SV=2, CEps=104.7955, QPEps=0.0000)
Iter 3: *(NumConst=3, SV=3, CEps=63.2759, QPEps=0.0000)
Iter 4: *(NumConst=4, SV=3, CEps=77.2612, QPEps=0.0000)
Iter 5: *(NumConst=5, SV=4, CEps=70.5662, QPEps=18.4394)
Iter 6: *(NumConst=6, SV=5, CEps=28.0911, QPEps=0.0000)
Iter 7: *(NumConst=7, SV=6, CEps=16.1296, QPEps=0.0000)
Iter 8: .........*(NumConst=8, SV=7, CEps=138.2599, QPEps=25.9832)
Iter 9: *(NumConst=9, SV=8, CEps=41.6624, QPEps=0.0000)
Iter 10: *(NumConst=10, SV=7, CEps=32.3581, QPEps=0.0000)
Iter 11: *(NumConst=11, SV=7, CEps=26.5346, QPEps=6.7941)
Iter 12: *(NumConst=12, SV=8, CEps=16.5669, QPEps=0.0000)
Iter 13: .........*(NumConst=13, SV=9, CEps=161.6170, QPEps=2.7990)
Iter 14: *(NumConst=14, SV=9, CEps=21.2484, QPEps=8.0122)
Iter 15: *(NumConst=15, SV=12, CEps=20.2454, QPEps=6.8460)
Iter 16: .........*(NumConst=16, SV=13, CEps=95.2769, QPEps=39.1840)
Iter 17: *(NumConst=17, SV=9, CEps=50.9596, QPEps=2.1292)
Iter 18: *(NumConst=18, SV=9, CEps=36.4190, QPEps=0.0000)
Iter 19: *(NumConst=19, SV=9, CEps=52.2944, QPEps=0.0000)
Iter 20: *(NumConst=20, SV=8, CEps=48.8463, QPEps=11.0787)
Iter 21: *(NumConst=21, SV=9, CEps=31.0304, QPEps=11.7882)
Iter 22: *(NumConst=22, SV=8, CEps=33.2940, QPEps=0.0000)
Iter 23: *(NumConst=23, SV=10, CEps=15.4884, QPEps=0.0000)
Iter 24: *(NumConst=24, SV=10, CEps=16.0322, QPEps=0.4107)
Iter 25: *(NumConst=25, SV=10, CEps=18.7593, QPEps=5.4317)
Iter 26: *(NumConst=26, SV=10, CEps=15.0529, QPEps=2.4498)
Iter 27: .........*(NumConst=27, SV=11, CEps=66.2117, QPEps=2.3478)
Iter 28: *(NumConst=28, SV=12, CEps=29.1440, QPEps=10.2963)
Iter 29: *(NumConst=29, SV=11, CEps=33.3890, QPEps=4.7510)
Iter 30: *(NumConst=30, SV=12, CEps=27.4683, QPEps=2.9208)
Iter 31: *(NumConst=31, SV=12, CEps=22.7823, QPEps=7.8955)
Iter 32: *(NumConst=32, SV=10, CEps=17.0925, QPEps=0.0000)
Iter 33: *(NumConst=33, SV=10, CEps=21.3136, QPEps=2.6378)
Iter 34: *(NumConst=34, SV=10, CEps=15.1880, QPEps=2.3639)
Iter 35: *(NumConst=35, SV=10, CEps=26.3341, QPEps=1.6664)
Iter 36: *(NumConst=36, SV=10, CEps=14.5593, QPEps=3.3525)
Iter 37: *(NumConst=37, SV=11, CEps=13.4301, QPEps=6.4258)
Iter 38: *(NumConst=38, SV=12, CEps=8.1152, QPEps=3.3439)
Iter 39: *(NumConst=39, SV=10, CEps=13.0774, QPEps=2.7159)
Iter 40: *(NumConst=40, SV=10, CEps=9.5853, QPEps=0.2531)
Iter 41: .........*(NumConst=41, SV=11, CEps=31.7813, QPEps=0.2880)
Iter 42: *(NumConst=42, SV=12, CEps=14.4834, QPEps=1.4063)
Iter 43: *(NumConst=43, SV=13, CEps=20.1238, QPEps=7.1756)
Iter 44: *(NumConst=44, SV=11, CEps=16.0730, QPEps=3.4776)
Iter 45: *(NumConst=45, SV=11, CEps=14.2782, QPEps=3.8174)
Iter 46: *(NumConst=46, SV=12, CEps=10.4246, QPEps=1.8309)
Iter 47: *(NumConst=47, SV=13, CEps=11.0156, QPEps=5.0376)
Iter 48: *(NumConst=48, SV=12, CEps=9.0679, QPEps=2.7649)
Iter 49: *(NumConst=49, SV=14, CEps=9.8612, QPEps=3.6115)
Iter 50: *(NumConst=50, SV=13, CEps=10.3187, QPEps=0.9381)
Iter 51: *(NumConst=51, SV=12, CEps=7.0678, QPEps=2.4136)
Iter 52: *(NumConst=52, SV=13, CEps=8.8238, QPEps=1.7352)
Iter 53: *(NumConst=53, SV=13, CEps=6.7634, QPEps=2.8398)
Iter 54: *(NumConst=54, SV=12, CEps=5.8112, QPEps=1.2947)
Iter 55: *(NumConst=55, SV=13, CEps=5.4570, QPEps=2.2522)
Iter 56: *(NumConst=56, SV=14, CEps=5.3976, QPEps=2.5131)
Iter 57: *(NumConst=57, SV=13, CEps=5.0187, QPEps=2.5025)
Iter 58: *(NumConst=58, SV=14, CEps=3.5837, QPEps=1.1939)
Iter 59: *(NumConst=59, SV=15, CEps=8.6570, QPEps=1.1547)
Iter 60: *(NumConst=59, SV=15, CEps=5.2535, QPEps=1.4338)
Iter 61: *(NumConst=59, SV=16, CEps=4.0943, QPEps=1.0989)
Iter 62: *(NumConst=60, SV=16, CEps=4.2089, QPEps=1.6825)
Iter 63: *(NumConst=60, SV=14, CEps=3.7336, QPEps=1.2152)
Iter 64: *(NumConst=61, SV=14, CEps=4.4802, QPEps=0.8560)
Iter 65: .........*(NumConst=62, SV=15, CEps=8.3189, QPEps=1.5710)
Iter 66: *(NumConst=60, SV=18, CEps=6.2696, QPEps=1.8598)
Iter 67: *(NumConst=59, SV=18, CEps=6.3714, QPEps=3.0154)
Iter 68: *(NumConst=60, SV=20, CEps=4.8317, QPEps=1.7619)
Iter 69: *(NumConst=60, SV=20, CEps=6.4077, QPEps=2.3513)
Iter 70: *(NumConst=60, SV=15, CEps=5.6277, QPEps=2.1219)
Iter 71: *(NumConst=59, SV=15, CEps=4.4709, QPEps=2.1173)
Iter 72: *(NumConst=60, SV=15, CEps=4.0522, QPEps=1.3510)
Iter 73: *(NumConst=61, SV=15, CEps=3.3043, QPEps=1.3816)
Iter 74: *(NumConst=61, SV=15, CEps=3.2555, QPEps=1.2715)
Iter 75: *(NumConst=61, SV=16, CEps=3.7148, QPEps=0.7257)
Iter 76: *(NumConst=62, SV=17, CEps=2.7057, QPEps=0.8962)
Iter 77: *(NumConst=62, SV=17, CEps=4.0293, QPEps=0.5981)
Iter 78: *(NumConst=61, SV=17, CEps=2.5990, QPEps=1.0653)
Iter 79: *(NumConst=61, SV=17, CEps=1.9961, QPEps=0.9518)
Iter 80: *(NumConst=61, SV=18, CEps=2.3744, QPEps=0.9541)
Iter 81: *(NumConst=60, SV=16, CEps=3.6413, QPEps=0.7378)
Iter 82: *(NumConst=60, SV=17, CEps=1.9638, QPEps=0.6763)
Iter 83: *(NumConst=59, SV=16, CEps=2.4536, QPEps=0.4335)
Iter 84: *(NumConst=60, SV=18, CEps=1.6814, QPEps=0.6923)
Iter 85: *(NumConst=60, SV=16, CEps=1.8362, QPEps=0.8285)
Iter 86: *(NumConst=61, SV=16, CEps=2.0868, QPEps=0.7981)
Iter 87: *(NumConst=62, SV=17, CEps=1.1217, QPEps=0.3319)
Iter 88: *(NumConst=61, SV=19, CEps=1.5390, QPEps=0.4680)
Iter 89: *(NumConst=61, SV=17, CEps=1.7757, QPEps=0.3442)
Iter 90: *(NumConst=62, SV=17, CEps=1.1250, QPEps=0.4545)
Iter 91: *(NumConst=62, SV=17, CEps=1.4292, QPEps=0.4218)
Iter 92: *(NumConst=63, SV=17, CEps=1.2545, QPEps=0.5577)
Iter 93: *(NumConst=63, SV=18, CEps=1.3879, QPEps=0.5276)
Iter 94: .........*(NumConst=63, SV=21, CEps=1.7518, QPEps=0.7853)
Iter 95: *(NumConst=64, SV=21, CEps=1.9739, QPEps=0.8089)
Iter 96: *(NumConst=64, SV=21, CEps=1.9305, QPEps=0.7046)
Iter 97: *(NumConst=64, SV=21, CEps=2.0301, QPEps=0.8343)
Iter 98: *(NumConst=64, SV=23, CEps=1.7560, QPEps=0.7566)
Iter 99: *(NumConst=62, SV=21, CEps=2.1369, QPEps=0.7968)
Iter 100: *(NumConst=61, SV=22, CEps=1.5225, QPEps=0.6651)
Iter 101: *(NumConst=62, SV=24, CEps=1.7384, QPEps=0.5969)
Iter 102: *(NumConst=62, SV=22, CEps=1.6146, QPEps=0.5651)
Iter 103: *(NumConst=62, SV=20, CEps=1.1258, QPEps=0.2980)
Iter 104: *(NumConst=63, SV=21, CEps=1.7359, QPEps=0.4698)
Iter 105: *(NumConst=64, SV=24, CEps=0.7737, QPEps=0.3547)
Iter 106: *(NumConst=63, SV=20, CEps=1.6990, QPEps=0.3486)
Iter 107: *(NumConst=64, SV=20, CEps=1.0751, QPEps=0.3607)
Iter 108: *(NumConst=65, SV=20, CEps=0.7818, QPEps=0.2882)
Iter 109: *(NumConst=66, SV=20, CEps=1.2399, QPEps=0.3224)
Iter 110: *(NumConst=66, SV=18, CEps=0.9687, QPEps=0.1925)
Iter 111: *(NumConst=66, SV=18, CEps=0.8146, QPEps=0.3174)
Iter 112: *(NumConst=67, SV=18, CEps=1.1285, QPEps=0.3218)
Iter 113: *(NumConst=68, SV=20, CEps=0.4912, QPEps=0.1852)
Iter 114: *(NumConst=69, SV=19, CEps=0.9799, QPEps=0.1839)
Iter 115: *(NumConst=70, SV=19, CEps=1.0752, QPEps=0.1887)
Iter 116: *(NumConst=71, SV=20, CEps=0.9945, QPEps=0.1895)
Iter 117: *(NumConst=71, SV=20, CEps=0.6489, QPEps=0.2453)
Iter 118: *(NumConst=72, SV=18, CEps=0.8070, QPEps=0.2023)
Iter 119: *(NumConst=67, SV=18, CEps=0.8873, QPEps=0.1414)
Iter 120: *(NumConst=67, SV=19, CEps=0.6805, QPEps=0.2338)
Iter 121: *(NumConst=66, SV=20, CEps=0.6033, QPEps=0.1363)
Iter 122: *(NumConst=66, SV=20, CEps=0.5758, QPEps=0.1779)
Iter 123: *(NumConst=66, SV=18, CEps=0.5600, QPEps=0.2237)
Iter 124: *(NumConst=66, SV=19, CEps=0.4662, QPEps=0.2038)
Iter 125: *(NumConst=67, SV=19, CEps=0.4514, QPEps=0.1614)
Iter 126: *(NumConst=67, SV=19, CEps=0.4009, QPEps=0.1025)
Iter 127: *(NumConst=67, SV=19, CEps=0.4681, QPEps=0.1871)
Iter 128: *(NumConst=67, SV=19, CEps=0.3728, QPEps=0.1749)
Iter 129: *(NumConst=67, SV=20, CEps=0.4733, QPEps=0.1812)
Iter 130: *(NumConst=66, SV=19, CEps=0.3453, QPEps=0.1299)
Iter 131: *(NumConst=67, SV=19, CEps=0.5184, QPEps=0.1636)
Iter 132: *(NumConst=67, SV=19, CEps=0.3496, QPEps=0.1719)
Iter 133: *(NumConst=68, SV=19, CEps=0.3755, QPEps=0.1180)
Iter 134: *(NumConst=67, SV=20, CEps=0.3558, QPEps=0.0917)
Iter 135: *(NumConst=68, SV=21, CEps=0.4609, QPEps=0.1717)
Iter 136: *(NumConst=68, SV=21, CEps=0.2998, QPEps=0.1363)
Iter 137: *(NumConst=69, SV=21, CEps=0.2992, QPEps=0.1155)
Iter 138: *(NumConst=68, SV=21, CEps=0.3923, QPEps=0.1397)
Iter 139: *(NumConst=68, SV=22, CEps=0.3078, QPEps=0.1371)
Iter 140: *(NumConst=68, SV=21, CEps=0.2960, QPEps=0.1140)
Iter 141: *(NumConst=69, SV=22, CEps=0.2773, QPEps=0.1105)
Iter 142: *(NumConst=70, SV=22, CEps=0.2383, QPEps=0.1148)
Iter 143: *(NumConst=71, SV=21, CEps=0.3105, QPEps=0.0997)
Iter 144: .........*(NumConst=71, SV=23, CEps=0.2459, QPEps=0.1175)
Iter 145: *(NumConst=72, SV=23, CEps=0.3519, QPEps=0.1112)
Iter 146: *(NumConst=73, SV=23, CEps=0.2800, QPEps=0.0816)
Iter 147: *(NumConst=74, SV=23, CEps=0.3018, QPEps=0.1088)
Iter 148: *(NumConst=72, SV=22, CEps=0.2942, QPEps=0.1026)
Iter 149: *(NumConst=73, SV=22, CEps=0.2446, QPEps=0.1003)
Iter 150: *(NumConst=74, SV=23, CEps=0.1931, QPEps=0.0852)
Iter 151: *(NumConst=75, SV=26, CEps=0.2565, QPEps=0.0798)
Iter 152: *(NumConst=74, SV=23, CEps=0.2021, QPEps=0.0878)
Iter 153: *(NumConst=75, SV=22, CEps=0.2734, QPEps=0.0741)
Iter 154: *(NumConst=76, SV=24, CEps=0.2237, QPEps=0.0866)
Iter 155: *(NumConst=73, SV=25, CEps=0.2050, QPEps=0.0849)
Iter 156: *(NumConst=73, SV=26, CEps=0.2442, QPEps=0.0897)
Iter 157: *(NumConst=71, SV=26, CEps=0.1554, QPEps=0.0755)
Iter 158: *(NumConst=71, SV=23, CEps=0.2076, QPEps=0.0598)
Iter 159: *(NumConst=69, SV=22, CEps=0.2232, QPEps=0.0568)
Iter 160: *(NumConst=69, SV=22, CEps=0.1909, QPEps=0.0550)
Iter 161: *(NumConst=70, SV=22, CEps=0.1617, QPEps=0.0662)
Iter 162: *(NumConst=70, SV=20, CEps=0.1502, QPEps=0.0488)
Iter 163: *(NumConst=68, SV=21, CEps=0.1554, QPEps=0.0547)
Iter 164: *(NumConst=68, SV=22, CEps=0.1441, QPEps=0.0645)
Iter 165: *(NumConst=69, SV=23, CEps=0.1065, QPEps=0.0487)
Iter 166: *(NumConst=70, SV=23, CEps=0.1352, QPEps=0.0457)
Iter 167: *(NumConst=69, SV=22, CEps=0.1353, QPEps=0.0491)
Iter 168: *(NumConst=69, SV=24, CEps=0.1167, QPEps=0.0490)
Iter 169: *(NumConst=70, SV=24, CEps=0.1489, QPEps=0.0504)
Iter 170: .........(NumConst=70, SV=24, CEps=0.0898, QPEps=0.0504)
Final epsilon on KKT-Conditions: 0.08984
Upper bound on duality gap: 9.18629
Dual objective value: dval=8522.33861
Primal objective value: pval=8531.52490
Total number of constraints in final working set: 70 (of 169)
Number of iterations: 170
Number of calls to 'find_most_violated_constraint': 48037
Number of SV: 24 
Norm of weight vector: |w|=52.28487
Value of slack variable (on working set): xi=71.56227
Value of slack variable (global): xi=71.64671
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=2.33436
Runtime in cpu-seconds: 21.84
Final number of constraints in cache: 21829
Compacting linear model...done
Writing learned model...done
=== END _tune-hyperparameter4: ./run learn ../cv.train --- OK [26s]
=== START _tune-hyperparameter4: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions4
Reading model...done.
Reading test examples... (1871 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.14
Average loss on test set: 11.8119
Zero/one-error on test set: 11.81% (1650 correct, 221 incorrect, 1871 total)
=== END _tune-hyperparameter4: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions4 --- OK [1s]
=== START program5: ./run evaluate ../program1/cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions4
=== END program5: ./run evaluate ../program1/cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions4 --- OK [1s]
CV error rate 0.11811865312667 with hyperparameter 100.0

Best hyperparameter value is 100.0; got CV error rate 0.11811865312667
=== END program1: ./run learn ../dataset6/train --- OK [78s]

===== MAIN: predict/evaluate on train data =====
=== START program7: ./run stripLabels ../dataset6/train ../program0/evalTrain.in
=== END program7: ./run stripLabels ../dataset6/train ../program0/evalTrain.in --- OK [2s]
=== START program1: ./run predict ../program0/evalTrain.in ../program0/evalTrain.out
=== START _tune-hyperparameter-best: ./run predict ../../program0/evalTrain.in ../../program0/evalTrain.out
Reading model...done.
Reading test examples... (6238 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.36
Average loss on test set: 96.3770
Zero/one-error on test set: 96.38% (226 correct, 6012 incorrect, 6238 total)
=== END _tune-hyperparameter-best: ./run predict ../../program0/evalTrain.in ../../program0/evalTrain.out --- OK [6s]
=== END program1: ./run predict ../program0/evalTrain.in ../program0/evalTrain.out --- OK [6s]
=== START program8: ./run evaluate ../dataset6/train ../program0/evalTrain.out
=== END program8: ./run evaluate ../dataset6/train ../program0/evalTrain.out --- OK [2s]

===== MAIN: predict/evaluate on test data =====
=== START program7: ./run stripLabels ../dataset6/test ../program0/evalTest.in
=== END program7: ./run stripLabels ../dataset6/test ../program0/evalTest.in --- OK [0s]
=== START program1: ./run predict ../program0/evalTest.in ../program0/evalTest.out
=== START _tune-hyperparameter-best: ./run predict ../../program0/evalTest.in ../../program0/evalTest.out
Reading model...done.
Reading test examples... (1559 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.13
Average loss on test set: 96.1514
Zero/one-error on test set: 96.15% (60 correct, 1499 incorrect, 1559 total)
=== END _tune-hyperparameter-best: ./run predict ../../program0/evalTest.in ../../program0/evalTest.out --- OK [2s]
=== END program1: ./run predict ../program0/evalTest.in ../program0/evalTest.out --- OK [2s]
=== START program8: ./run evaluate ../dataset6/test ../program0/evalTest.out
=== END program8: ./run evaluate ../dataset6/test ../program0/evalTest.out --- OK [0s]


real	1m31.210s
user	1m27.037s
sys	0m3.248s

Run specification

• supervised-learning: Main entry for supervised learning for training and testing a program on a dataset.
  • (learner:Program) tune-hyperparameter: Sets the hyperparameter
    • (numProbes:int) 5
    • (learner:Program) svmlight_multiclass-linear: SVMlight for multiclass classification (http://svmlight.joachims.org/svm_multiclass.html)
    • (splitter:Program) multiclass-utils: Validates and inspects a dataset in MulticlassClassification format.
    • (evaluator:Program[Evaluate]) classification-evaluator: Evaluates predictions of classification datasets (discrete outputs).
  • (dataset:Dataset) isolet: 7797 examples, 617 features
  • (stripper:Program[Strip]) multiclass-utils: Validates and inspects a dataset in MulticlassClassification format.
  • (evaluator:Program[Evaluate]) classification-evaluator: Evaluates predictions of classification datasets (discrete outputs).

Results

• doTest:
  • evaluate:
    • errorRate: 0.131494547787043
    • numErrors: 205
    • numExamples: 1559
    • success: true
    • time: 0
  • predict:
    • predict:
      • success: true
      • time: 2
    • success: true
    • time: 2
  • strip:
    • success: true
    • time: 0
• doTrain:
  • evaluate:
    • errorRate: 0.108207758897082
    • numErrors: 675
    • numExamples: 6238
    • success: true
    • time: 2
  • predict:
    • predict:
      • success: true
      • time: 6
    • success: true
    • time: 6
  • strip:
    • success: true
    • time: 2
• exitCode: 0
• learn:
  • bestCVErrorRate: 0.11811865312667
  • bestHyperparameter: 100.0
  • evaluate0:
    • errorRate: 0.666488508818813
    • numErrors: 1247
    • numExamples: 1871
    • success: true
    • time: 0
  • evaluate1:
    • errorRate: 0.679315873864244
    • numErrors: 1271
    • numExamples: 1871
    • success: true
    • time: 0
  • evaluate2:
    • errorRate: 0.298236237306253
    • numErrors: 558
    • numExamples: 1871
    • success: true
    • time: 1
  • evaluate3:
    • errorRate: 0.140566541956173
    • numErrors: 263
    • numExamples: 1871
    • success: true
    • time: 0
  • evaluate4:
    • errorRate: 0.11811865312667
    • numErrors: 221
    • numExamples: 1871
    • success: true
    • time: 1
  • learn0:
    • success: true
    • time: 5
  • learn1:
    • success: true
    • time: 8
  • learn2:
    • success: true
    • time: 10
  • learn3:
    • success: true
    • time: 16
  • learn4:
    • success: true
    • time: 26
  • predict0:
    • success: true
    • time: 2
  • predict1:
    • success: true
    • time: 2
  • predict2:
    • success: true
    • time: 2
  • predict3:
    • success: true
    • time: 2
  • predict4:
    • success: true
    • time: 1
  • setHyperparameter0:
    • success: true
    • time: 0
  • setHyperparameter1:
    • success: true
    • time: 0
  • setHyperparameter2:
    • success: true
    • time: 0
  • setHyperparameter3:
    • success: true
    • time: 0
  • setHyperparameter4:
    • success: true
    • time: 0
  • split:
    • success: true
    • time: 2
  • success: true
  • time: 78
• success: true

Comments: