MLcomp version alpha Statistics: 2046 users, 665 datasets, 389 programs, 18920 runs (0 queued, 4 running), 0 workers

Run 13211

Creator	yorinasub17
Program	svmlight_multiclass
Dataset	mnistSmall
Task type	MulticlassClassification
Created	1y364d ago
Download	Login required!

Status: Done!

Total Time	9s
Max Memory Usage	69M
Domain	MulticlassClassification
Learn time
Train error	0
Predict train time
Test error	0.147
Predict test time

Log file

===== MAIN: learn based on training data =====
=== START program1: ./run learn ../dataset2/train
Using hyperparameter c = 0.1
Reading training examples... (700 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=19.4135, QPEps=0.0000)
Iter 3: .........*(NumConst=3, SV=3, CEps=109.2702, QPEps=0.0000)
Iter 4: *(NumConst=4, SV=3, CEps=51.9005, QPEps=0.0000)
Iter 5: *(NumConst=5, SV=4, CEps=43.0679, QPEps=0.0000)
Iter 6: *(NumConst=6, SV=5, CEps=36.4901, QPEps=0.0000)
Iter 7: *(NumConst=7, SV=6, CEps=32.9307, QPEps=0.0000)
Iter 8: *(NumConst=8, SV=5, CEps=41.2458, QPEps=0.0000)
Iter 9: *(NumConst=9, SV=7, CEps=29.6210, QPEps=0.0000)
Iter 10: .........*(NumConst=10, SV=9, CEps=143.7445, QPEps=28.3862)
Iter 11: *(NumConst=11, SV=9, CEps=45.5933, QPEps=6.8849)
Iter 12: *(NumConst=12, SV=8, CEps=67.6884, QPEps=0.0000)
Iter 13: *(NumConst=13, SV=8, CEps=28.9548, QPEps=0.0000)
Iter 14: *(NumConst=14, SV=9, CEps=42.7847, QPEps=12.3956)
Iter 15: *(NumConst=15, SV=9, CEps=34.0332, QPEps=13.1021)
Iter 16: *(NumConst=16, SV=10, CEps=18.7934, QPEps=0.0000)
Iter 17: *(NumConst=17, SV=10, CEps=22.5481, QPEps=4.3731)
Iter 18: *(NumConst=18, SV=10, CEps=16.8412, QPEps=0.2314)
Iter 19: *(NumConst=19, SV=10, CEps=16.0361, QPEps=3.4116)
Iter 20: .........*(NumConst=20, SV=11, CEps=109.6733, QPEps=1.5199)
Iter 21: *(NumConst=21, SV=11, CEps=26.7336, QPEps=7.9013)
Iter 22: *(NumConst=22, SV=13, CEps=31.8479, QPEps=5.7802)
Iter 23: *(NumConst=23, SV=13, CEps=26.4551, QPEps=10.9228)
Iter 24: *(NumConst=24, SV=13, CEps=16.2485, QPEps=7.3704)
Iter 25: *(NumConst=25, SV=13, CEps=18.5291, QPEps=1.4281)
Iter 26: *(NumConst=26, SV=15, CEps=15.5888, QPEps=6.1397)
Iter 27: *(NumConst=27, SV=15, CEps=16.3925, QPEps=5.3510)
Iter 28: *(NumConst=28, SV=17, CEps=15.7123, QPEps=4.9407)
Iter 29: *(NumConst=29, SV=15, CEps=17.4072, QPEps=5.3130)
Iter 30: .........*(NumConst=30, SV=15, CEps=31.4159, QPEps=13.8534)
Iter 31: *(NumConst=31, SV=14, CEps=24.5212, QPEps=5.2125)
Iter 32: *(NumConst=32, SV=16, CEps=19.2873, QPEps=8.1218)
Iter 33: *(NumConst=33, SV=16, CEps=21.3831, QPEps=9.2790)
Iter 34: *(NumConst=34, SV=17, CEps=20.2180, QPEps=4.5984)
Iter 35: *(NumConst=35, SV=18, CEps=14.2522, QPEps=5.7199)
Iter 36: *(NumConst=36, SV=20, CEps=14.5839, QPEps=6.7645)
Iter 37: *(NumConst=37, SV=20, CEps=13.4048, QPEps=6.5736)
Iter 38: *(NumConst=38, SV=20, CEps=13.0675, QPEps=3.9534)
Iter 39: *(NumConst=39, SV=21, CEps=13.7997, QPEps=5.6171)
Iter 40: *(NumConst=40, SV=21, CEps=12.0230, QPEps=5.1755)
Iter 41: *(NumConst=41, SV=22, CEps=11.8230, QPEps=5.8896)
Iter 42: *(NumConst=42, SV=22, CEps=9.8522, QPEps=4.3706)
Iter 43: *(NumConst=43, SV=23, CEps=10.2874, QPEps=4.2150)
Iter 44: *(NumConst=44, SV=24, CEps=10.1660, QPEps=4.4148)
Iter 45: *(NumConst=45, SV=24, CEps=10.0603, QPEps=3.9767)
Iter 46: *(NumConst=46, SV=25, CEps=9.6161, QPEps=4.7162)
Iter 47: *(NumConst=47, SV=26, CEps=7.3239, QPEps=3.4273)
Iter 48: *(NumConst=48, SV=26, CEps=9.1287, QPEps=3.5898)
Iter 49: *(NumConst=49, SV=27, CEps=6.6812, QPEps=2.3070)
Iter 50: *(NumConst=50, SV=28, CEps=6.7176, QPEps=3.1342)
Iter 51: *(NumConst=51, SV=28, CEps=7.6585, QPEps=3.1070)
Iter 52: *(NumConst=52, SV=28, CEps=5.4556, QPEps=2.7241)
Iter 53: *(NumConst=52, SV=29, CEps=5.8462, QPEps=2.5452)
Iter 54: *(NumConst=53, SV=32, CEps=4.9563, QPEps=2.0456)
Iter 55: *(NumConst=54, SV=33, CEps=4.9203, QPEps=2.3743)
Iter 56: *(NumConst=55, SV=33, CEps=4.5405, QPEps=2.2449)
Iter 57: *(NumConst=56, SV=34, CEps=4.2153, QPEps=1.9918)
Iter 58: *(NumConst=57, SV=34, CEps=3.9887, QPEps=1.9053)
Iter 59: .........*(NumConst=58, SV=35, CEps=8.1249, QPEps=3.5654)
Iter 60: *(NumConst=58, SV=36, CEps=7.4471, QPEps=2.9740)
Iter 61: *(NumConst=57, SV=36, CEps=5.3773, QPEps=2.4753)
Iter 62: *(NumConst=57, SV=37, CEps=6.4749, QPEps=2.6659)
Iter 63: *(NumConst=58, SV=38, CEps=4.4237, QPEps=2.1979)
Iter 64: *(NumConst=58, SV=37, CEps=4.8674, QPEps=1.6511)
Iter 65: *(NumConst=59, SV=38, CEps=4.4918, QPEps=1.7524)
Iter 66: *(NumConst=59, SV=38, CEps=3.7608, QPEps=1.1436)
Iter 67: *(NumConst=60, SV=38, CEps=3.7706, QPEps=1.7268)
Iter 68: *(NumConst=60, SV=39, CEps=3.3756, QPEps=1.4862)
Iter 69: *(NumConst=61, SV=40, CEps=2.8740, QPEps=1.3946)
Iter 70: *(NumConst=62, SV=40, CEps=3.2619, QPEps=1.4215)
Iter 71: *(NumConst=63, SV=42, CEps=2.7992, QPEps=1.2588)
Iter 72: *(NumConst=63, SV=42, CEps=3.8614, QPEps=1.2370)
Iter 73: *(NumConst=64, SV=41, CEps=3.1486, QPEps=1.0341)
Iter 74: *(NumConst=65, SV=42, CEps=2.8329, QPEps=1.2783)
Iter 75: *(NumConst=66, SV=43, CEps=2.9657, QPEps=0.9924)
Iter 76: *(NumConst=66, SV=44, CEps=1.9210, QPEps=0.9344)
Iter 77: *(NumConst=67, SV=44, CEps=2.6321, QPEps=0.9594)
Iter 78: *(NumConst=66, SV=44, CEps=2.4197, QPEps=0.8593)
Iter 79: *(NumConst=67, SV=44, CEps=2.3666, QPEps=0.9224)
Iter 80: *(NumConst=67, SV=45, CEps=2.2432, QPEps=0.9556)
Iter 81: *(NumConst=68, SV=46, CEps=2.0438, QPEps=0.8906)
Iter 82: *(NumConst=67, SV=47, CEps=1.8132, QPEps=0.8905)
Iter 83: *(NumConst=68, SV=49, CEps=1.5518, QPEps=0.7260)
Iter 84: *(NumConst=69, SV=49, CEps=1.5897, QPEps=0.7268)
Iter 85: *(NumConst=69, SV=49, CEps=1.6101, QPEps=0.6554)
Iter 86: *(NumConst=70, SV=49, CEps=1.8037, QPEps=0.7589)
Iter 87: *(NumConst=70, SV=48, CEps=1.1626, QPEps=0.5540)
Iter 88: *(NumConst=70, SV=48, CEps=1.3005, QPEps=0.5033)
Iter 89: *(NumConst=71, SV=48, CEps=1.4578, QPEps=0.5074)
Iter 90: *(NumConst=72, SV=48, CEps=1.2260, QPEps=0.5583)
Iter 91: *(NumConst=73, SV=49, CEps=1.0963, QPEps=0.4684)
Iter 92: *(NumConst=74, SV=50, CEps=0.9209, QPEps=0.4579)
Iter 93: *(NumConst=74, SV=51, CEps=1.1410, QPEps=0.4592)
Iter 94: *(NumConst=73, SV=51, CEps=0.9672, QPEps=0.4518)
Iter 95: *(NumConst=74, SV=52, CEps=0.8575, QPEps=0.4075)
Iter 96: *(NumConst=75, SV=53, CEps=0.9478, QPEps=0.4197)
Iter 97: *(NumConst=76, SV=53, CEps=0.9058, QPEps=0.3366)
Iter 98: *(NumConst=76, SV=55, CEps=0.9210, QPEps=0.4163)
Iter 99: .........*(NumConst=77, SV=57, CEps=1.6678, QPEps=0.7579)
Iter 100: *(NumConst=78, SV=57, CEps=1.6456, QPEps=0.8214)
Iter 101: *(NumConst=79, SV=57, CEps=1.4257, QPEps=0.6465)
Iter 102: *(NumConst=80, SV=57, CEps=1.6090, QPEps=0.6737)
Iter 103: *(NumConst=81, SV=58, CEps=1.3689, QPEps=0.6827)
Iter 104: *(NumConst=82, SV=59, CEps=1.1942, QPEps=0.5150)
Iter 105: *(NumConst=82, SV=60, CEps=1.0759, QPEps=0.4890)
Iter 106: *(NumConst=83, SV=61, CEps=1.0993, QPEps=0.5037)
Iter 107: *(NumConst=83, SV=61, CEps=1.0248, QPEps=0.4280)
Iter 108: *(NumConst=84, SV=62, CEps=0.9574, QPEps=0.4584)
Iter 109: *(NumConst=85, SV=63, CEps=0.9557, QPEps=0.4364)
Iter 110: *(NumConst=85, SV=63, CEps=0.8438, QPEps=0.4189)
Iter 111: *(NumConst=86, SV=64, CEps=0.8797, QPEps=0.3914)
Iter 112: *(NumConst=87, SV=65, CEps=0.7118, QPEps=0.3193)
Iter 113: *(NumConst=86, SV=66, CEps=0.6699, QPEps=0.2805)
Iter 114: *(NumConst=87, SV=67, CEps=0.6722, QPEps=0.3327)
Iter 115: *(NumConst=87, SV=66, CEps=0.5773, QPEps=0.2872)
Iter 116: *(NumConst=87, SV=67, CEps=0.6700, QPEps=0.2783)
Iter 117: *(NumConst=88, SV=68, CEps=0.6312, QPEps=0.2712)
Iter 118: *(NumConst=89, SV=68, CEps=0.6012, QPEps=0.2846)
Iter 119: *(NumConst=90, SV=69, CEps=0.7070, QPEps=0.2846)
Iter 120: *(NumConst=91, SV=68, CEps=0.5240, QPEps=0.2523)
Iter 121: *(NumConst=91, SV=69, CEps=0.5334, QPEps=0.2481)
Iter 122: *(NumConst=92, SV=70, CEps=0.5404, QPEps=0.2558)
Iter 123: *(NumConst=93, SV=71, CEps=0.5039, QPEps=0.2035)
Iter 124: *(NumConst=94, SV=72, CEps=0.4869, QPEps=0.2104)
Iter 125: *(NumConst=95, SV=73, CEps=0.4600, QPEps=0.2064)
Iter 126: *(NumConst=95, SV=74, CEps=0.4313, QPEps=0.1997)
Iter 127: *(NumConst=95, SV=74, CEps=0.4643, QPEps=0.2091)
Iter 128: *(NumConst=95, SV=75, CEps=0.4242, QPEps=0.2062)
Iter 129: *(NumConst=96, SV=75, CEps=0.4416, QPEps=0.1806)
Iter 130: *(NumConst=97, SV=75, CEps=0.3515, QPEps=0.1590)
Iter 131: *(NumConst=98, SV=76, CEps=0.3546, QPEps=0.1488)
Iter 132: *(NumConst=99, SV=77, CEps=0.3747, QPEps=0.1502)
Iter 133: *(NumConst=99, SV=78, CEps=0.3315, QPEps=0.1654)
Iter 134: *(NumConst=99, SV=79, CEps=0.3019, QPEps=0.1368)
Iter 135: *(NumConst=100, SV=78, CEps=0.3547, QPEps=0.1242)
Iter 136: *(NumConst=99, SV=79, CEps=0.3315, QPEps=0.1462)
Iter 137: *(NumConst=99, SV=79, CEps=0.3550, QPEps=0.1503)
Iter 138: *(NumConst=99, SV=80, CEps=0.2591, QPEps=0.1245)
Iter 139: *(NumConst=99, SV=81, CEps=0.2982, QPEps=0.1224)
Iter 140: *(NumConst=100, SV=82, CEps=0.2842, QPEps=0.1257)
Iter 141: *(NumConst=101, SV=83, CEps=0.2744, QPEps=0.1279)
Iter 142: *(NumConst=102, SV=83, CEps=0.2834, QPEps=0.1289)
Iter 143: *(NumConst=102, SV=84, CEps=0.2610, QPEps=0.1268)
Iter 144: *(NumConst=103, SV=85, CEps=0.2601, QPEps=0.1262)
Iter 145: *(NumConst=104, SV=85, CEps=0.2931, QPEps=0.1100)
Iter 146: *(NumConst=104, SV=86, CEps=0.2629, QPEps=0.1266)
Iter 147: *(NumConst=105, SV=86, CEps=0.2399, QPEps=0.1171)
Iter 148: *(NumConst=106, SV=87, CEps=0.2243, QPEps=0.0960)
Iter 149: *(NumConst=106, SV=88, CEps=0.2533, QPEps=0.1113)
Iter 150: *(NumConst=106, SV=88, CEps=0.2309, QPEps=0.1071)
Iter 151: *(NumConst=106, SV=89, CEps=0.2075, QPEps=0.0820)
Iter 152: *(NumConst=107, SV=89, CEps=0.1925, QPEps=0.0802)
Iter 153: *(NumConst=108, SV=90, CEps=0.2165, QPEps=0.0937)
Iter 154: *(NumConst=109, SV=91, CEps=0.1816, QPEps=0.0857)
Iter 155: *(NumConst=110, SV=91, CEps=0.1916, QPEps=0.0861)
Iter 156: .........*(NumConst=110, SV=92, CEps=0.3299, QPEps=0.1628)
Iter 157: *(NumConst=111, SV=93, CEps=0.3105, QPEps=0.1481)
Iter 158: *(NumConst=112, SV=94, CEps=0.3087, QPEps=0.1513)
Iter 159: *(NumConst=112, SV=94, CEps=0.3023, QPEps=0.1498)
Iter 160: *(NumConst=113, SV=95, CEps=0.2799, QPEps=0.1367)
Iter 161: *(NumConst=114, SV=96, CEps=0.2396, QPEps=0.1196)
Iter 162: *(NumConst=115, SV=97, CEps=0.2734, QPEps=0.1188)
Iter 163: *(NumConst=116, SV=98, CEps=0.2908, QPEps=0.1159)
Iter 164: *(NumConst=115, SV=99, CEps=0.2586, QPEps=0.1158)
Iter 165: *(NumConst=116, SV=100, CEps=0.3129, QPEps=0.1084)
Iter 166: *(NumConst=117, SV=101, CEps=0.2297, QPEps=0.1129)
Iter 167: *(NumConst=117, SV=102, CEps=0.2469, QPEps=0.1141)
Iter 168: *(NumConst=118, SV=103, CEps=0.2354, QPEps=0.1132)
Iter 169: *(NumConst=117, SV=104, CEps=0.2272, QPEps=0.1016)
Iter 170: *(NumConst=118, SV=105, CEps=0.2207, QPEps=0.0988)
Iter 171: *(NumConst=119, SV=106, CEps=0.2083, QPEps=0.1002)
Iter 172: *(NumConst=120, SV=107, CEps=0.2173, QPEps=0.1028)
Iter 173: *(NumConst=121, SV=108, CEps=0.2268, QPEps=0.0971)
Iter 174: *(NumConst=122, SV=109, CEps=0.1989, QPEps=0.0920)
Iter 175: *(NumConst=123, SV=110, CEps=0.1717, QPEps=0.0807)
Iter 176: *(NumConst=123, SV=111, CEps=0.1602, QPEps=0.0775)
Iter 177: *(NumConst=124, SV=112, CEps=0.2081, QPEps=0.0773)
Iter 178: *(NumConst=124, SV=113, CEps=0.1741, QPEps=0.0770)
Iter 179: *(NumConst=124, SV=113, CEps=0.1725, QPEps=0.0775)
Iter 180: *(NumConst=125, SV=114, CEps=0.1751, QPEps=0.0701)
Iter 181: *(NumConst=126, SV=114, CEps=0.1477, QPEps=0.0728)
Iter 182: *(NumConst=127, SV=115, CEps=0.1602, QPEps=0.0729)
Iter 183: *(NumConst=128, SV=116, CEps=0.1438, QPEps=0.0711)
Iter 184: *(NumConst=127, SV=117, CEps=0.1422, QPEps=0.0657)
Iter 185: *(NumConst=128, SV=118, CEps=0.1323, QPEps=0.0659)
Iter 186: *(NumConst=128, SV=118, CEps=0.1346, QPEps=0.0603)
Iter 187: *(NumConst=129, SV=119, CEps=0.1185, QPEps=0.0576)
Iter 188: *(NumConst=130, SV=119, CEps=0.1107, QPEps=0.0513)
Iter 189: *(NumConst=131, SV=120, CEps=0.1090, QPEps=0.0524)
Iter 190: *(NumConst=132, SV=119, CEps=0.1055, QPEps=0.0467)
Iter 191: *(NumConst=132, SV=120, CEps=0.1191, QPEps=0.0522)
Iter 192: *(NumConst=133, SV=119, CEps=0.1158, QPEps=0.0525)
Iter 193: *(NumConst=134, SV=120, CEps=0.1102, QPEps=0.0454)
Iter 194: .........(NumConst=134, SV=120, CEps=0.0978, QPEps=0.0454)
Final epsilon on KKT-Conditions: 0.09784
Upper bound on duality gap: 0.00932
Dual objective value: dval=1.14084
Primal objective value: pval=1.15015
Total number of constraints in final working set: 134 (of 193)
Number of iterations: 194
Number of calls to 'find_most_violated_constraint': 6300
Number of SV: 120 
Norm of weight vector: |w|=1.39857
Value of slack variable (on working set): xi=1.65172
Value of slack variable (global): xi=1.72157
Norm of longest difference vector: ||Psi(x,y)-Psi(x,ybar)||=318.61281
Runtime in cpu-seconds: 2.80
Final number of constraints in cache: 3385
Compacting linear model...done
Writing learned model...done
=== END program1: ./run learn ../dataset2/train --- OK [9s]

===== MAIN: predict/evaluate on train data =====
=== START program3: ./run stripLabels ../dataset2/train ../program0/evalTrain.in
=== END program3: ./run stripLabels ../dataset2/train ../program0/evalTrain.in --- OK [1s]
=== START program1: ./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: 0.04
Average loss on test set: 90.0000
Zero/one-error on test set: 90.00% (70 correct, 630 incorrect, 700 total)
=== END program1: ./run predict ../program0/evalTrain.in ../program0/evalTrain.out --- OK [1s]
=== START program4: ./run evaluate ../dataset2/train ../program0/evalTrain.out
=== END program4: ./run evaluate ../dataset2/train ../program0/evalTrain.out --- OK [1s]

===== MAIN: predict/evaluate on test data =====
=== START program3: ./run stripLabels ../dataset2/test ../program0/evalTest.in
=== END program3: ./run stripLabels ../dataset2/test ../program0/evalTest.in --- OK [0s]
=== START program1: ./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: 89.6667
Zero/one-error on test set: 89.67% (31 correct, 269 incorrect, 300 total)
=== END program1: ./run predict ../program0/evalTest.in ../program0/evalTest.out --- OK [1s]
=== START program4: ./run evaluate ../dataset2/test ../program0/evalTest.out
=== END program4: ./run evaluate ../dataset2/test ../program0/evalTest.out --- OK [1s]


real	0m13.474s
user	0m5.216s
sys	0m0.352s

Run specification

• supervised-learning: Main entry for supervised learning for training and testing a program on a dataset.
  • (learner:Program) svmlight_multiclass: SVMlight for multiclass classification (http://svmlight.joachims.org/svm_multiclass.html)
  • (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: 0.146666666666667
    • numErrors: 44
    • numExamples: 300
    • success: true
    • time: 1
  • predict:
    • success: true
    • time: 1
  • strip:
    • success: true
    • time: 0
• doTrain:
  • evaluate:
    • errorRate: 0.0
    • numErrors: 0
    • numExamples: 700
    • success: true
    • time: 1
  • predict:
    • success: true
    • time: 1
  • strip:
    • success: true
    • time: 1
• exitCode: 0
• learn:
  • success: true
  • time: 9
• success: true

Comments: