MLcomp version alpha Statistics: 1213 users, 402 datasets, 328 programs, 9938 runs (0 queued, 0 running), 0 workers

Run 7271

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

Status: Done!

Total Time	15s
Max Memory Usage	65M
Domain:	MulticlassClassification
Learn time
Train error	0.076
Predict train time
Test error	0.140
Predict test time

Log file

... (lines omitted) ...
Iter 79: *(NumConst=70, SV=47, CEps=1.0174, QPEps=0.4742)
Iter 80: *(NumConst=71, SV=46, CEps=1.1383, QPEps=0.4382)
Iter 81: *(NumConst=71, SV=47, CEps=1.0452, QPEps=0.3889)
Iter 82: *(NumConst=71, SV=49, CEps=1.0151, QPEps=0.5030)
Iter 83: *(NumConst=72, SV=49, CEps=1.1296, QPEps=0.5023)
Iter 84: .........*(NumConst=73, SV=49, CEps=2.6057, QPEps=1.0867)
Iter 85: *(NumConst=71, SV=50, CEps=2.2010, QPEps=1.0424)
Iter 86: *(NumConst=71, SV=50, CEps=1.6815, QPEps=0.6999)
Iter 87: *(NumConst=72, SV=53, CEps=1.6201, QPEps=0.7772)
Iter 88: *(NumConst=72, SV=54, CEps=1.8096, QPEps=0.6462)
Iter 89: *(NumConst=73, SV=54, CEps=1.3878, QPEps=0.6757)
Iter 90: *(NumConst=74, SV=55, CEps=1.1225, QPEps=0.5489)
Iter 91: *(NumConst=74, SV=54, CEps=1.1190, QPEps=0.4042)
Iter 92: *(NumConst=73, SV=54, CEps=1.1090, QPEps=0.5093)
Iter 93: *(NumConst=74, SV=55, CEps=1.1465, QPEps=0.5154)
Iter 94: *(NumConst=75, SV=56, CEps=0.9769, QPEps=0.4673)
Iter 95: *(NumConst=76, SV=57, CEps=0.8265, QPEps=0.3513)
Iter 96: *(NumConst=77, SV=58, CEps=0.7400, QPEps=0.3649)
Iter 97: *(NumConst=77, SV=59, CEps=0.8273, QPEps=0.3377)
Iter 98: *(NumConst=77, SV=60, CEps=0.7903, QPEps=0.2800)
Iter 99: *(NumConst=78, SV=61, CEps=0.8096, QPEps=0.3691)
Iter 100: *(NumConst=79, SV=62, CEps=0.6360, QPEps=0.2865)
Iter 101: *(NumConst=79, SV=63, CEps=0.6778, QPEps=0.3123)
Iter 102: *(NumConst=78, SV=64, CEps=0.6044, QPEps=0.2845)
Iter 103: *(NumConst=79, SV=65, CEps=0.5336, QPEps=0.2385)
Iter 104: *(NumConst=80, SV=66, CEps=0.5769, QPEps=0.2055)
Iter 105: *(NumConst=81, SV=67, CEps=0.5189, QPEps=0.1975)
Iter 106: *(NumConst=80, SV=68, CEps=0.6256, QPEps=0.2437)
Iter 107: *(NumConst=81, SV=69, CEps=0.4798, QPEps=0.2159)
Iter 108: *(NumConst=82, SV=70, CEps=0.5845, QPEps=0.1748)
Iter 109: *(NumConst=83, SV=69, CEps=0.4632, QPEps=0.1830)
Iter 110: *(NumConst=84, SV=70, CEps=0.4686, QPEps=0.2177)
Iter 111: *(NumConst=84, SV=71, CEps=0.4459, QPEps=0.1881)
Iter 112: *(NumConst=85, SV=72, CEps=0.3945, QPEps=0.1869)
Iter 113: *(NumConst=85, SV=73, CEps=0.4119, QPEps=0.1889)
Iter 114: *(NumConst=85, SV=74, CEps=0.4098, QPEps=0.1859)
Iter 115: *(NumConst=86, SV=76, CEps=0.3619, QPEps=0.1632)
Iter 116: *(NumConst=87, SV=77, CEps=0.3594, QPEps=0.1764)
Iter 117: *(NumConst=88, SV=78, CEps=0.4573, QPEps=0.1542)
Iter 118: *(NumConst=89, SV=79, CEps=0.3064, QPEps=0.1303)
Iter 119: *(NumConst=90, SV=80, CEps=0.3294, QPEps=0.1318)
Iter 120: *(NumConst=91, SV=81, CEps=0.3525, QPEps=0.1350)
Iter 121: *(NumConst=92, SV=82, CEps=0.3715, QPEps=0.1419)
Iter 122: *(NumConst=93, SV=82, CEps=0.3724, QPEps=0.1364)
Iter 123: *(NumConst=94, SV=83, CEps=0.2653, QPEps=0.1310)
Iter 124: *(NumConst=94, SV=84, CEps=0.2692, QPEps=0.1251)
Iter 125: *(NumConst=95, SV=85, CEps=0.3301, QPEps=0.1325)
Iter 126: *(NumConst=96, SV=86, CEps=0.2980, QPEps=0.1238)
Iter 127: *(NumConst=97, SV=87, CEps=0.2744, QPEps=0.1182)
Iter 128: .........*(NumConst=97, SV=88, CEps=0.5313, QPEps=0.2594)
Iter 129: *(NumConst=97, SV=89, CEps=0.5331, QPEps=0.2585)
Iter 130: *(NumConst=98, SV=90, CEps=0.5108, QPEps=0.2502)
Iter 131: *(NumConst=99, SV=91, CEps=0.4257, QPEps=0.2102)
Iter 132: *(NumConst=99, SV=92, CEps=0.3828, QPEps=0.1786)
Iter 133: *(NumConst=99, SV=93, CEps=0.3530, QPEps=0.1736)
Iter 134: *(NumConst=100, SV=94, CEps=0.3756, QPEps=0.1718)
Iter 135: *(NumConst=101, SV=95, CEps=0.3264, QPEps=0.1622)
Iter 136: *(NumConst=102, SV=94, CEps=0.3359, QPEps=0.1436)
Iter 137: *(NumConst=103, SV=94, CEps=0.2569, QPEps=0.1282)
Iter 138: *(NumConst=104, SV=95, CEps=0.2818, QPEps=0.1160)
Iter 139: *(NumConst=104, SV=95, CEps=0.2775, QPEps=0.1230)
Iter 140: *(NumConst=103, SV=96, CEps=0.2530, QPEps=0.1166)
Iter 141: *(NumConst=103, SV=96, CEps=0.2825, QPEps=0.1206)
Iter 142: *(NumConst=104, SV=97, CEps=0.2340, QPEps=0.1170)
Iter 143: *(NumConst=105, SV=97, CEps=0.2393, QPEps=0.1134)
Iter 144: *(NumConst=106, SV=98, CEps=0.2139, QPEps=0.0942)
Iter 145: *(NumConst=107, SV=98, CEps=0.2673, QPEps=0.1036)
Iter 146: *(NumConst=108, SV=99, CEps=0.2292, QPEps=0.1068)
Iter 147: *(NumConst=109, SV=100, CEps=0.2044, QPEps=0.1011)
Iter 148: *(NumConst=110, SV=101, CEps=0.1926, QPEps=0.0943)
Iter 149: *(NumConst=111, SV=102, CEps=0.2213, QPEps=0.0909)
Iter 150: *(NumConst=112, SV=103, CEps=0.1857, QPEps=0.0874)
Iter 151: *(NumConst=113, SV=104, CEps=0.1916, QPEps=0.0928)
Iter 152: *(NumConst=114, SV=105, CEps=0.1862, QPEps=0.0796)
Iter 153: *(NumConst=115, SV=105, CEps=0.1652, QPEps=0.0804)
Iter 154: *(NumConst=116, SV=106, CEps=0.1834, QPEps=0.0818)
Iter 155: *(NumConst=117, SV=107, CEps=0.1821, QPEps=0.0691)
Iter 156: *(NumConst=118, SV=108, CEps=0.1779, QPEps=0.0774)
Iter 157: *(NumConst=119, SV=109, CEps=0.1618, QPEps=0.0805)
Iter 158: *(NumConst=119, SV=110, CEps=0.1306, QPEps=0.0642)
Iter 159: *(NumConst=120, SV=111, CEps=0.1516, QPEps=0.0632)
Iter 160: *(NumConst=121, SV=112, CEps=0.1383, QPEps=0.0644)
Iter 161: *(NumConst=122, SV=113, CEps=0.1335, QPEps=0.0579)
Iter 162: *(NumConst=123, SV=114, CEps=0.1493, QPEps=0.0638)
Iter 163: *(NumConst=124, SV=115, CEps=0.1424, QPEps=0.0645)
Iter 164: *(NumConst=125, SV=116, CEps=0.1297, QPEps=0.0612)
Iter 165: *(NumConst=126, SV=117, CEps=0.1440, QPEps=0.0612)
Iter 166: *(NumConst=127, SV=117, CEps=0.1307, QPEps=0.0581)
Iter 167: *(NumConst=128, SV=118, CEps=0.1244, QPEps=0.0572)
Iter 168: *(NumConst=129, SV=119, CEps=0.1282, QPEps=0.0597)
Iter 169: *(NumConst=130, SV=120, CEps=0.1098, QPEps=0.0547)
Iter 170: *(NumConst=131, SV=121, CEps=0.1313, QPEps=0.0532)
Iter 171: *(NumConst=131, SV=122, CEps=0.1096, QPEps=0.0497)
Iter 172: *(NumConst=132, SV=123, CEps=0.1208, QPEps=0.0495)
Iter 173: *(NumConst=133, SV=124, CEps=0.1140, QPEps=0.0498)
Iter 174: *(NumConst=134, SV=125, CEps=0.1177, QPEps=0.0485)
Iter 175: *(NumConst=135, SV=126, CEps=0.1094, QPEps=0.0493)
Iter 176: *(NumConst=136, SV=126, CEps=0.1055, QPEps=0.0528)
Iter 177: *(NumConst=137, SV=127, CEps=0.1208, QPEps=0.0520)
Iter 178: .........*(NumConst=138, SV=128, CEps=0.1289, QPEps=0.0637)
Iter 179: *(NumConst=139, SV=129, CEps=0.1469, QPEps=0.0614)
Iter 180: *(NumConst=140, SV=130, CEps=0.1238, QPEps=0.0499)
Iter 181: *(NumConst=141, SV=131, CEps=0.1109, QPEps=0.0509)
Iter 182: *(NumConst=142, SV=131, CEps=0.1065, QPEps=0.0490)
Iter 183: *(NumConst=143, SV=132, CEps=0.1202, QPEps=0.0516)
Iter 184: *(NumConst=144, SV=133, CEps=0.1093, QPEps=0.0520)
Iter 185: *(NumConst=143, SV=134, CEps=0.1047, QPEps=0.0515)
Iter 186: *(NumConst=143, SV=135, CEps=0.1006, QPEps=0.0482)
Iter 187: .........(NumConst=143, SV=135, CEps=0.0957, QPEps=0.0482)
Final epsilon on KKT-Conditions: 0.09572
Upper bound on duality gap: 0.95713
Dual objective value: dval=0.71798
Primal objective value: pval=1.67511
Total number of constraints in final working set: 143 (of 186)
Number of iterations: 187
Number of calls to 'find_most_violated_constraint': 4900
Number of SV: 135 
Norm of weight vector: |w|=1.19827
Value of slack variable (on working set): xi=0.02669
Value of slack variable (global): xi=0.09572
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=1271.59068
Runtime in cpu-seconds: 2.42
Final number of constraints in cache: 2436
Compacting linear model...done
Writing learned model...done
=== END _tune-hyperparameter3: ./run learn ../cv.train --- OK [2s]
=== START _tune-hyperparameter3: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions3
Reading model...done.
Reading test examples... (210 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.00
Average loss on test set: 18.0952
Zero/one-error on test set: 18.10% (172 correct, 38 incorrect, 210 total)
=== END _tune-hyperparameter3: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions3 --- OK [1s]
=== 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 0.180952380952381 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... (490 examples) done
Training set properties: 784 features, 10 classes
Iter 1: .........*(NumConst=1, SV=1, CEps=100.0000, QPEps=0.0000)
Iter 2: *(NumConst=2, SV=2, CEps=19.6128, QPEps=0.0000)
Iter 3: .........*(NumConst=3, SV=3, CEps=109.7147, QPEps=0.0000)
Iter 4: *(NumConst=4, SV=4, CEps=51.8742, QPEps=0.0000)
Iter 5: *(NumConst=5, SV=4, CEps=44.3148, QPEps=0.0000)
Iter 6: *(NumConst=6, SV=5, CEps=23.7262, QPEps=0.0000)
Iter 7: *(NumConst=7, SV=6, CEps=35.9320, QPEps=0.0000)
Iter 8: *(NumConst=8, SV=5, CEps=38.7229, QPEps=0.0000)
Iter 9: *(NumConst=9, SV=7, CEps=26.3414, QPEps=0.0000)
Iter 10: .........*(NumConst=10, SV=9, CEps=131.9045, QPEps=0.0000)
Iter 11: *(NumConst=11, SV=9, CEps=39.5320, QPEps=0.0000)
Iter 12: *(NumConst=12, SV=9, CEps=43.2025, QPEps=0.0000)
Iter 13: *(NumConst=13, SV=10, CEps=22.1083, QPEps=0.0000)
Iter 14: *(NumConst=14, SV=11, CEps=32.9500, QPEps=0.0000)
Iter 15: *(NumConst=15, SV=12, CEps=14.2774, QPEps=1.6306)
Iter 16: *(NumConst=16, SV=12, CEps=22.9430, QPEps=6.6001)
Iter 17: *(NumConst=17, SV=13, CEps=14.5043, QPEps=2.1719)
Iter 18: .........*(NumConst=18, SV=14, CEps=124.3741, QPEps=8.8984)
Iter 19: *(NumConst=19, SV=14, CEps=20.1195, QPEps=3.3750)
Iter 20: *(NumConst=20, SV=16, CEps=23.4966, QPEps=7.2218)
Iter 21: *(NumConst=21, SV=16, CEps=31.5525, QPEps=9.2431)
Iter 22: .........*(NumConst=22, SV=17, CEps=44.1610, QPEps=20.8151)
Iter 23: *(NumConst=23, SV=18, CEps=32.0830, QPEps=14.0636)
Iter 24: *(NumConst=24, SV=18, CEps=29.6314, QPEps=13.5201)
Iter 25: *(NumConst=25, SV=18, CEps=33.0510, QPEps=11.1203)
Iter 26: *(NumConst=26, SV=18, CEps=20.9366, QPEps=7.6502)
Iter 27: *(NumConst=27, SV=19, CEps=20.4102, QPEps=8.0710)
Iter 28: *(NumConst=28, SV=18, CEps=22.4809, QPEps=9.3438)
Iter 29: *(NumConst=29, SV=19, CEps=14.2537, QPEps=5.3360)
Iter 30: *(NumConst=30, SV=19, CEps=27.0904, QPEps=4.4046)
Iter 31: *(NumConst=31, SV=21, CEps=16.2152, QPEps=7.0177)
Iter 32: *(NumConst=32, SV=22, CEps=20.4327, QPEps=4.9027)
Iter 33: *(NumConst=33, SV=21, CEps=17.1271, QPEps=2.5550)
Iter 34: *(NumConst=34, SV=22, CEps=11.0623, QPEps=4.7211)
Iter 35: *(NumConst=35, SV=24, CEps=10.9006, QPEps=3.9681)
Iter 36: *(NumConst=36, SV=20, CEps=12.0781, QPEps=1.7289)
Iter 37: *(NumConst=37, SV=21, CEps=10.9375, QPEps=4.1646)
Iter 38: *(NumConst=38, SV=21, CEps=11.9545, QPEps=5.1445)
Iter 39: *(NumConst=39, SV=21, CEps=7.7789, QPEps=2.1604)
Iter 40: *(NumConst=40, SV=22, CEps=6.9723, QPEps=3.4054)
Iter 41: *(NumConst=41, SV=24, CEps=7.2996, QPEps=2.2282)
Iter 42: *(NumConst=42, SV=24, CEps=9.2590, QPEps=3.3925)
Iter 43: *(NumConst=43, SV=23, CEps=7.4707, QPEps=2.1362)
Iter 44: *(NumConst=44, SV=24, CEps=6.5067, QPEps=3.2394)
Iter 45: *(NumConst=45, SV=25, CEps=5.2703, QPEps=1.9908)
Iter 46: *(NumConst=46, SV=27, CEps=6.5143, QPEps=2.4108)
Iter 47: *(NumConst=47, SV=28, CEps=4.7762, QPEps=1.8845)
Iter 48: .........*(NumConst=48, SV=28, CEps=9.0548, QPEps=4.1055)
Iter 49: *(NumConst=49, SV=28, CEps=7.3511, QPEps=1.9314)
Iter 50: *(NumConst=50, SV=28, CEps=7.3017, QPEps=3.4533)
Iter 51: *(NumConst=51, SV=30, CEps=5.9597, QPEps=2.7967)
Iter 52: *(NumConst=52, SV=30, CEps=5.5330, QPEps=2.7617)
Iter 53: *(NumConst=53, SV=29, CEps=4.8327, QPEps=2.3776)
Iter 54: *(NumConst=54, SV=30, CEps=5.1415, QPEps=2.2759)
Iter 55: *(NumConst=55, SV=31, CEps=4.0006, QPEps=1.9361)
Iter 56: *(NumConst=56, SV=32, CEps=5.5775, QPEps=1.7870)
Iter 57: *(NumConst=56, SV=31, CEps=3.6680, QPEps=1.7603)
Iter 58: *(NumConst=57, SV=32, CEps=4.0647, QPEps=1.8226)
Iter 59: *(NumConst=58, SV=33, CEps=2.8968, QPEps=1.3485)
Iter 60: *(NumConst=58, SV=34, CEps=3.2491, QPEps=1.1457)
Iter 61: *(NumConst=58, SV=34, CEps=2.6159, QPEps=1.1616)
Iter 62: *(NumConst=59, SV=35, CEps=2.8624, QPEps=1.2236)
Iter 63: *(NumConst=60, SV=36, CEps=2.5063, QPEps=1.2156)
Iter 64: *(NumConst=61, SV=36, CEps=2.7498, QPEps=1.0551)
Iter 65: *(NumConst=61, SV=36, CEps=2.0200, QPEps=1.0089)
Iter 66: *(NumConst=62, SV=37, CEps=1.7680, QPEps=0.7156)
Iter 67: *(NumConst=63, SV=38, CEps=2.4880, QPEps=0.8157)
Iter 68: *(NumConst=64, SV=39, CEps=2.2515, QPEps=0.8204)
Iter 69: *(NumConst=65, SV=40, CEps=2.1786, QPEps=0.8771)
Iter 70: *(NumConst=65, SV=41, CEps=1.3776, QPEps=0.6659)
Iter 71: *(NumConst=66, SV=42, CEps=1.7851, QPEps=0.6297)
Iter 72: *(NumConst=67, SV=43, CEps=1.7549, QPEps=0.4726)
Iter 73: *(NumConst=68, SV=43, CEps=1.5412, QPEps=0.6795)
Iter 74: *(NumConst=69, SV=44, CEps=1.4208, QPEps=0.6424)
Iter 75: *(NumConst=69, SV=44, CEps=1.2757, QPEps=0.6003)
Iter 76: *(NumConst=70, SV=45, CEps=1.3748, QPEps=0.5965)
Iter 77: *(NumConst=69, SV=46, CEps=1.3770, QPEps=0.5789)
Iter 78: *(NumConst=69, SV=47, CEps=1.2368, QPEps=0.5389)
Iter 79: *(NumConst=70, SV=47, CEps=1.0174, QPEps=0.4742)
Iter 80: *(NumConst=71, SV=46, CEps=1.1383, QPEps=0.4382)
Iter 81: *(NumConst=71, SV=47, CEps=1.0452, QPEps=0.3889)
Iter 82: *(NumConst=71, SV=49, CEps=1.0151, QPEps=0.5030)
Iter 83: *(NumConst=72, SV=49, CEps=1.1296, QPEps=0.5023)
Iter 84: .........*(NumConst=73, SV=49, CEps=2.6057, QPEps=1.0867)
Iter 85: *(NumConst=71, SV=50, CEps=2.2010, QPEps=1.0424)
Iter 86: *(NumConst=71, SV=50, CEps=1.6815, QPEps=0.6999)
Iter 87: *(NumConst=72, SV=53, CEps=1.6201, QPEps=0.7772)
Iter 88: *(NumConst=72, SV=54, CEps=1.8096, QPEps=0.6462)
Iter 89: *(NumConst=73, SV=54, CEps=1.3878, QPEps=0.6757)
Iter 90: *(NumConst=74, SV=55, CEps=1.1225, QPEps=0.5489)
Iter 91: *(NumConst=74, SV=54, CEps=1.1190, QPEps=0.4042)
Iter 92: *(NumConst=73, SV=54, CEps=1.1090, QPEps=0.5093)
Iter 93: *(NumConst=74, SV=55, CEps=1.1465, QPEps=0.5154)
Iter 94: *(NumConst=75, SV=56, CEps=0.9769, QPEps=0.4673)
Iter 95: *(NumConst=76, SV=57, CEps=0.8265, QPEps=0.3513)
Iter 96: *(NumConst=77, SV=58, CEps=0.7400, QPEps=0.3649)
Iter 97: *(NumConst=77, SV=59, CEps=0.8273, QPEps=0.3377)
Iter 98: *(NumConst=77, SV=60, CEps=0.7903, QPEps=0.2800)
Iter 99: *(NumConst=78, SV=61, CEps=0.8096, QPEps=0.3691)
Iter 100: *(NumConst=79, SV=62, CEps=0.6360, QPEps=0.2865)
Iter 101: *(NumConst=79, SV=63, CEps=0.6778, QPEps=0.3123)
Iter 102: *(NumConst=78, SV=64, CEps=0.6044, QPEps=0.2845)
Iter 103: *(NumConst=79, SV=65, CEps=0.5336, QPEps=0.2385)
Iter 104: *(NumConst=80, SV=66, CEps=0.5769, QPEps=0.2055)
Iter 105: *(NumConst=81, SV=67, CEps=0.5189, QPEps=0.1975)
Iter 106: *(NumConst=80, SV=68, CEps=0.6256, QPEps=0.2437)
Iter 107: *(NumConst=81, SV=69, CEps=0.4798, QPEps=0.2159)
Iter 108: *(NumConst=82, SV=70, CEps=0.5845, QPEps=0.1748)
Iter 109: *(NumConst=83, SV=69, CEps=0.4632, QPEps=0.1830)
Iter 110: *(NumConst=84, SV=70, CEps=0.4686, QPEps=0.2177)
Iter 111: *(NumConst=84, SV=71, CEps=0.4459, QPEps=0.1881)
Iter 112: *(NumConst=85, SV=72, CEps=0.3945, QPEps=0.1869)
Iter 113: *(NumConst=85, SV=73, CEps=0.4119, QPEps=0.1889)
Iter 114: *(NumConst=85, SV=74, CEps=0.4098, QPEps=0.1859)
Iter 115: *(NumConst=86, SV=76, CEps=0.3619, QPEps=0.1632)
Iter 116: *(NumConst=87, SV=77, CEps=0.3594, QPEps=0.1764)
Iter 117: *(NumConst=88, SV=78, CEps=0.4573, QPEps=0.1542)
Iter 118: *(NumConst=89, SV=79, CEps=0.3064, QPEps=0.1303)
Iter 119: *(NumConst=90, SV=80, CEps=0.3294, QPEps=0.1318)
Iter 120: *(NumConst=91, SV=81, CEps=0.3525, QPEps=0.1350)
Iter 121: *(NumConst=92, SV=82, CEps=0.3715, QPEps=0.1419)
Iter 122: *(NumConst=93, SV=82, CEps=0.3724, QPEps=0.1364)
Iter 123: *(NumConst=94, SV=83, CEps=0.2653, QPEps=0.1310)
Iter 124: *(NumConst=94, SV=84, CEps=0.2692, QPEps=0.1251)
Iter 125: *(NumConst=95, SV=85, CEps=0.3301, QPEps=0.1325)
Iter 126: *(NumConst=96, SV=86, CEps=0.2980, QPEps=0.1238)
Iter 127: *(NumConst=97, SV=87, CEps=0.2744, QPEps=0.1182)
Iter 128: .........*(NumConst=97, SV=88, CEps=0.5313, QPEps=0.2594)
Iter 129: *(NumConst=97, SV=89, CEps=0.5331, QPEps=0.2585)
Iter 130: *(NumConst=98, SV=90, CEps=0.5108, QPEps=0.2502)
Iter 131: *(NumConst=99, SV=91, CEps=0.4257, QPEps=0.2102)
Iter 132: *(NumConst=99, SV=92, CEps=0.3828, QPEps=0.1786)
Iter 133: *(NumConst=99, SV=93, CEps=0.3530, QPEps=0.1736)
Iter 134: *(NumConst=100, SV=94, CEps=0.3756, QPEps=0.1718)
Iter 135: *(NumConst=101, SV=95, CEps=0.3264, QPEps=0.1622)
Iter 136: *(NumConst=102, SV=94, CEps=0.3359, QPEps=0.1436)
Iter 137: *(NumConst=103, SV=94, CEps=0.2569, QPEps=0.1282)
Iter 138: *(NumConst=104, SV=95, CEps=0.2818, QPEps=0.1160)
Iter 139: *(NumConst=104, SV=95, CEps=0.2775, QPEps=0.1230)
Iter 140: *(NumConst=103, SV=96, CEps=0.2530, QPEps=0.1166)
Iter 141: *(NumConst=103, SV=96, CEps=0.2825, QPEps=0.1206)
Iter 142: *(NumConst=104, SV=97, CEps=0.2340, QPEps=0.1170)
Iter 143: *(NumConst=105, SV=97, CEps=0.2393, QPEps=0.1134)
Iter 144: *(NumConst=106, SV=98, CEps=0.2139, QPEps=0.0942)
Iter 145: *(NumConst=107, SV=98, CEps=0.2673, QPEps=0.1036)
Iter 146: *(NumConst=108, SV=99, CEps=0.2292, QPEps=0.1068)
Iter 147: *(NumConst=109, SV=100, CEps=0.2044, QPEps=0.1011)
Iter 148: *(NumConst=110, SV=101, CEps=0.1926, QPEps=0.0943)
Iter 149: *(NumConst=111, SV=102, CEps=0.2213, QPEps=0.0909)
Iter 150: *(NumConst=112, SV=103, CEps=0.1857, QPEps=0.0874)
Iter 151: *(NumConst=113, SV=104, CEps=0.1916, QPEps=0.0928)
Iter 152: *(NumConst=114, SV=105, CEps=0.1862, QPEps=0.0796)
Iter 153: *(NumConst=115, SV=105, CEps=0.1652, QPEps=0.0804)
Iter 154: *(NumConst=116, SV=106, CEps=0.1834, QPEps=0.0818)
Iter 155: *(NumConst=117, SV=107, CEps=0.1821, QPEps=0.0691)
Iter 156: *(NumConst=118, SV=108, CEps=0.1779, QPEps=0.0774)
Iter 157: *(NumConst=119, SV=109, CEps=0.1618, QPEps=0.0805)
Iter 158: *(NumConst=119, SV=110, CEps=0.1306, QPEps=0.0642)
Iter 159: *(NumConst=120, SV=111, CEps=0.1516, QPEps=0.0632)
Iter 160: *(NumConst=121, SV=112, CEps=0.1383, QPEps=0.0644)
Iter 161: *(NumConst=122, SV=113, CEps=0.1335, QPEps=0.0579)
Iter 162: *(NumConst=123, SV=114, CEps=0.1493, QPEps=0.0638)
Iter 163: *(NumConst=124, SV=115, CEps=0.1424, QPEps=0.0645)
Iter 164: *(NumConst=125, SV=116, CEps=0.1297, QPEps=0.0612)
Iter 165: *(NumConst=126, SV=117, CEps=0.1440, QPEps=0.0612)
Iter 166: *(NumConst=127, SV=117, CEps=0.1307, QPEps=0.0581)
Iter 167: *(NumConst=128, SV=118, CEps=0.1244, QPEps=0.0572)
Iter 168: *(NumConst=129, SV=119, CEps=0.1282, QPEps=0.0597)
Iter 169: *(NumConst=130, SV=120, CEps=0.1098, QPEps=0.0547)
Iter 170: *(NumConst=131, SV=121, CEps=0.1313, QPEps=0.0532)
Iter 171: *(NumConst=131, SV=122, CEps=0.1096, QPEps=0.0497)
Iter 172: *(NumConst=132, SV=123, CEps=0.1208, QPEps=0.0495)
Iter 173: *(NumConst=133, SV=124, CEps=0.1140, QPEps=0.0498)
Iter 174: *(NumConst=134, SV=125, CEps=0.1177, QPEps=0.0485)
Iter 175: *(NumConst=135, SV=126, CEps=0.1094, QPEps=0.0493)
Iter 176: *(NumConst=136, SV=126, CEps=0.1055, QPEps=0.0528)
Iter 177: *(NumConst=137, SV=127, CEps=0.1208, QPEps=0.0520)
Iter 178: .........*(NumConst=138, SV=128, CEps=0.1289, QPEps=0.0637)
Iter 179: *(NumConst=139, SV=129, CEps=0.1469, QPEps=0.0614)
Iter 180: *(NumConst=140, SV=130, CEps=0.1238, QPEps=0.0499)
Iter 181: *(NumConst=141, SV=131, CEps=0.1109, QPEps=0.0509)
Iter 182: *(NumConst=142, SV=131, CEps=0.1065, QPEps=0.0490)
Iter 183: *(NumConst=143, SV=132, CEps=0.1202, QPEps=0.0516)
Iter 184: *(NumConst=144, SV=133, CEps=0.1093, QPEps=0.0520)
Iter 185: *(NumConst=143, SV=134, CEps=0.1047, QPEps=0.0515)
Iter 186: *(NumConst=143, SV=135, CEps=0.1006, QPEps=0.0482)
Iter 187: .........(NumConst=143, SV=135, CEps=0.0957, QPEps=0.0482)
Final epsilon on KKT-Conditions: 0.09572
Upper bound on duality gap: 9.57178
Dual objective value: dval=0.71798
Primal objective value: pval=10.28975
Total number of constraints in final working set: 143 (of 186)
Number of iterations: 187
Number of calls to 'find_most_violated_constraint': 4900
Number of SV: 135 
Norm of weight vector: |w|=1.19827
Value of slack variable (on working set): xi=0.02669
Value of slack variable (global): xi=0.09572
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=1271.59068
Runtime in cpu-seconds: 2.47
Final number of constraints in cache: 2436
Compacting linear model...done
Writing learned model...done
=== END _tune-hyperparameter4: ./run learn ../cv.train --- OK [3s]
=== START _tune-hyperparameter4: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions4
Reading model...done.
Reading test examples... (210 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.02
Average loss on test set: 18.0952
Zero/one-error on test set: 18.10% (172 correct, 38 incorrect, 210 total)
=== END _tune-hyperparameter4: ./run predict ../cv.test /home/mlcomp/worker1/scratch/program1/cvTestPredictions4 --- OK [0s]
=== 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 [0s]
CV error rate 0.180952380952381 with hyperparameter 100.0

Best hyperparameter value is 0.01; got CV error rate 0.152380952380952
=== END program1: ./run learn ../dataset6/train --- OK [16s]

===== 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 [1s]
=== 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... (700 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.02
Average loss on test set: 89.5714
Zero/one-error on test set: 89.57% (73 correct, 627 incorrect, 700 total)
=== END _tune-hyperparameter-best: ./run predict ../../program0/evalTrain.in ../../program0/evalTrain.out --- OK [0s]
=== END program1: ./run predict ../program0/evalTrain.in ../program0/evalTrain.out --- OK [0s]
=== START program8: ./run evaluate ../dataset6/train ../program0/evalTrain.out
=== END program8: ./run evaluate ../dataset6/train ../program0/evalTrain.out --- OK [1s]

===== 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... (300 examples) done.
Classifying test examples...done
Runtime (without IO) in cpu-seconds: 0.01
Average loss on test set: 89.3333
Zero/one-error on test set: 89.33% (32 correct, 268 incorrect, 300 total)
=== END _tune-hyperparameter-best: ./run predict ../../program0/evalTest.in ../../program0/evalTest.out --- OK [0s]
=== END program1: ./run predict ../program0/evalTest.in ../program0/evalTest.out --- OK [0s]
=== START program8: ./run evaluate ../dataset6/test ../program0/evalTest.out
=== END program8: ./run evaluate ../dataset6/test ../program0/evalTest.out --- OK [0s]


real	0m18.640s
user	0m16.985s
sys	0m1.012s

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) mnistSmall: MNIST handwritten digits.
  • (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: 0.14
    • numErrors: 42
    • numExamples: 300
    • success: true
    • time: 0
  • predict:
    • predict:
      • success: true
      • time: 0
    • success: true
    • time: 0
  • strip:
    • success: true
    • time: 0
• doTrain:
  • evaluate:
    • errorRate: 0.0757142857142857
    • numErrors: 53
    • numExamples: 700
    • success: true
    • time: 1
  • predict:
    • predict:
      • success: true
      • time: 0
    • success: true
    • time: 0
  • strip:
    • success: true
    • time: 1
• exitCode: 0
• learn:
  • bestCVErrorRate: 0.152380952380952
  • bestHyperparameter: 0.01
  • evaluate0:
    • errorRate: 0.152380952380952
    • numErrors: 32
    • numExamples: 210
    • success: true
    • time: 0
  • evaluate1:
    • errorRate: 0.176190476190476
    • numErrors: 37
    • numExamples: 210
    • success: true
    • time: 0
  • evaluate2:
    • errorRate: 0.180952380952381
    • numErrors: 38
    • numExamples: 210
    • success: true
    • time: 1
  • evaluate3:
    • errorRate: 0.180952380952381
    • numErrors: 38
    • numExamples: 210
    • success: true
    • time: 0
  • evaluate4:
    • errorRate: 0.180952380952381
    • numErrors: 38
    • numExamples: 210
    • success: true
    • time: 0
  • learn0:
    • success: true
    • time: 2
  • learn1:
    • success: true
    • time: 3
  • learn2:
    • success: true
    • time: 3
  • learn3:
    • success: true
    • time: 2
  • learn4:
    • success: true
    • time: 3
  • predict0:
    • success: true
    • time: 0
  • predict1:
    • success: true
    • time: 0
  • predict2:
    • success: true
    • time: 0
  • predict3:
    • success: true
    • time: 1
  • predict4:
    • success: true
    • time: 0
  • 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: 1
  • success: true
  • time: 16
• success: true

Comments: