copying files to /scratch...
starting benchmark...
/scratch/knn/venv/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters
running only kgraph
order: [Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 4, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 80, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 1, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 50, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 20, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 100, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 30, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 5, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 90, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 40, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 60, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 2, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 10, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 70, {'reverse': -1}, False]), Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 3, {'reverse': -1}, False])]
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 4, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 4, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.008 accuracy: 1.6488 cost: 0.00633344 M: 10 delta: 1 time: 0.613556 one-recall: 0 one-ratio: 1.98824
iteration: 2 recall: 0.0748 accuracy: 0.576643 cost: 0.0102207 M: 10 delta: 0.893264 time: 0.85509 one-recall: 0.07 one-ratio: 1.46524
iteration: 3 recall: 0.4584 accuracy: 0.129751 cost: 0.0167282 M: 11.1226 delta: 0.845956 time: 1.17848 one-recall: 0.46 one-ratio: 1.12263
iteration: 4 recall: 0.914 accuracy: 0.00790804 cost: 0.0248724 M: 11.72 delta: 0.566024 time: 1.55261 one-recall: 0.97 one-ratio: 1.006
iteration: 5 recall: 0.9892 accuracy: 0.000422819 cost: 0.0376455 M: 17.4207 delta: 0.223985 time: 2.09897 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9932 accuracy: 0.000213504 cost: 0.0459803 M: 21.1667 delta: 0.133705 time: 2.48822 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 46.45
Index size:  97448.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0043016667
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0068228210, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Reject!
1844.4 < 1902.54
  -> Decision False in time 0.1200000000, query time of that 0.0408971390, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
2335.36 < 2459.12
  -> Decision False in time 0.2200000000, query time of that 0.0768314640, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0082802960, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Reject!
1978.51 < 1981.25
  -> Decision False in time 0.5800000000, query time of that 0.0364480740, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
2706.99 < 2742.45
  -> Decision False in time 1.3500000000, query time of that 0.0802061250, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.4000000000, query time of that 0.0098229300, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1250.89 < 1424.99
  -> Decision False in time 4.1100000000, query time of that 0.0295231530, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2199.8 < 2320.86
  -> Decision False in time 10.7700000000, query time of that 0.0762243900, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 80, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 80, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0056 accuracy: 1.82173 cost: 0.00633344 M: 10 delta: 1 time: 6.87912 one-recall: 0 one-ratio: 1.99459
iteration: 2 recall: 0.0724 accuracy: 0.596222 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5051 one-recall: 0.11 one-ratio: 1.41587
iteration: 3 recall: 0.4572 accuracy: 0.132143 cost: 0.0167507 M: 11.1153 delta: 0.845787 time: 15.5432 one-recall: 0.47 one-ratio: 1.09055
iteration: 4 recall: 0.9232 accuracy: 0.00818741 cost: 0.0249114 M: 11.725 delta: 0.566225 time: 21.5066 one-recall: 0.93 one-ratio: 1.00497
iteration: 5 recall: 0.9892 accuracy: 0.000422568 cost: 0.0376859 M: 17.423 delta: 0.224572 time: 30.3852 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9952 accuracy: 0.000152319 cost: 0.046014 M: 21.1552 delta: 0.13416 time: 36.0959 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.41
Index size:  99516.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0004150000
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0300000000, query time of that 0.0109442160, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2300000000, query time of that 0.1164250680, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 2.3300000000, query time of that 1.1101867840, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1400000000, query time of that 0.0130704330, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.3300000000, query time of that 0.1209588850, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Accept!
  -> Decision True in time 13.6600000000, query time of that 1.2395076120, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0139069920, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Accept!
  -> Decision True in time 13.4600000000, query time of that 0.1413186650, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Accept!
  -> Decision True in time 136.3900000000, query time of that 1.4003953350, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 1, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 1, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0028 accuracy: 1.87063 cost: 0.00633344 M: 10 delta: 1 time: 6.87804 one-recall: 0 one-ratio: 1.91043
iteration: 2 recall: 0.0636 accuracy: 0.635218 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5012 one-recall: 0.09 one-ratio: 1.35214
iteration: 3 recall: 0.4656 accuracy: 0.131685 cost: 0.0167507 M: 11.1153 delta: 0.845796 time: 15.5372 one-recall: 0.6 one-ratio: 1.07085
iteration: 4 recall: 0.9188 accuracy: 0.00819028 cost: 0.0249109 M: 11.7244 delta: 0.566222 time: 21.4989 one-recall: 0.96 one-ratio: 1.00356
iteration: 5 recall: 0.9868 accuracy: 0.000705696 cost: 0.037684 M: 17.4233 delta: 0.224573 time: 30.3774 one-recall: 0.99 one-ratio: 1.00156
iteration: 6 recall: 0.9956 accuracy: 0.000123157 cost: 0.0460189 M: 21.1568 delta: 0.134156 time: 36.0974 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.43000000000001
Index size:  99528.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0422000000
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0055506760, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Reject!
2260.76 < 2956.12
  -> Decision False in time 0.0100000000, query time of that 0.0029879760, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
2236.97 < 3016.45
  -> Decision False in time 0.0000000000, query time of that 0.0010942010, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0057199730, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Reject!
2259.95 < 2765.63
  -> Decision False in time 0.0100000000, query time of that 0.0004923040, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
1054.79 < 1159.41
  -> Decision False in time 0.3200000000, query time of that 0.0146112580, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0068906460, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
2405.82 < 2773.19
  -> Decision False in time 5.7500000000, query time of that 0.0301905580, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1749.83 < 2842.2
  -> Decision False in time 0.6200000000, query time of that 0.0032284610, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 50, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 50, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0064 accuracy: 1.66291 cost: 0.00633344 M: 10 delta: 1 time: 6.87629 one-recall: 0 one-ratio: 1.78958
iteration: 2 recall: 0.0664 accuracy: 0.562034 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.4998 one-recall: 0.08 one-ratio: 1.33637
iteration: 3 recall: 0.4552 accuracy: 0.115564 cost: 0.0167507 M: 11.1153 delta: 0.845794 time: 15.5358 one-recall: 0.52 one-ratio: 1.07607
iteration: 4 recall: 0.920799 accuracy: 0.00714396 cost: 0.0249112 M: 11.7248 delta: 0.566233 time: 21.4954 one-recall: 0.96 one-ratio: 1.00671
iteration: 5 recall: 0.984 accuracy: 0.000770266 cost: 0.0376872 M: 17.4232 delta: 0.224524 time: 30.3703 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9924 accuracy: 0.000311837 cost: 0.0460242 M: 21.1571 delta: 0.134126 time: 36.0844 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.420000000000016
Index size:  99512.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0007133333
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0091640030, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2100000000, query time of that 0.0871900640, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 2.0900000000, query time of that 0.8677224510, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1400000000, query time of that 0.0096192050, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Reject!
1809.03 < 1871.82
  -> Decision False in time 0.4700000000, query time of that 0.0358698190, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
2143.56 < 2364.83
  -> Decision False in time 9.0900000000, query time of that 0.6699348240, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0113322000, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Accept!
  -> Decision True in time 13.4800000000, query time of that 0.1105052220, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1504.06 < 1533.62
  -> Decision False in time 7.3100000000, query time of that 0.0588893800, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 20, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 20, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0084 accuracy: 1.64044 cost: 0.00633344 M: 10 delta: 1 time: 6.8768 one-recall: 0.02 one-ratio: 1.72366
iteration: 2 recall: 0.0788 accuracy: 0.491113 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5002 one-recall: 0.08 one-ratio: 1.34378
iteration: 3 recall: 0.4904 accuracy: 0.102605 cost: 0.0167507 M: 11.1153 delta: 0.845796 time: 15.5354 one-recall: 0.56 one-ratio: 1.09779
iteration: 4 recall: 0.918 accuracy: 0.00753096 cost: 0.0249117 M: 11.7247 delta: 0.56622 time: 21.4956 one-recall: 0.95 one-ratio: 1.00965
iteration: 5 recall: 0.9836 accuracy: 0.00158116 cost: 0.0376819 M: 17.4219 delta: 0.224584 time: 30.3686 one-recall: 0.99 one-ratio: 1.001
iteration: 6 recall: 0.9928 accuracy: 0.000303187 cost: 0.0460308 M: 21.1614 delta: 0.134057 time: 36.0929 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.430000000000064
Index size:  99520.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0016866667
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0061299490, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.1800000000, query time of that 0.0603960770, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 1.8100000000, query time of that 0.5894422920, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0070116660, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.2700000000, query time of that 0.0677398390, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
2499.32 < 2516.12
  -> Decision False in time 0.9900000000, query time of that 0.0530040730, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3800000000, query time of that 0.0075243270, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1079.74 < 1189.22
  -> Decision False in time 9.0800000000, query time of that 0.0542311470, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1316.32 < 1390.14
  -> Decision False in time 10.6700000000, query time of that 0.0631335120, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 100, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 100, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0064 accuracy: 1.66231 cost: 0.00633344 M: 10 delta: 1 time: 6.88091 one-recall: 0 one-ratio: 1.86603
iteration: 2 recall: 0.0692 accuracy: 0.559591 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5045 one-recall: 0.03 one-ratio: 1.36931
iteration: 3 recall: 0.4824 accuracy: 0.113345 cost: 0.0167507 M: 11.1153 delta: 0.845788 time: 15.5392 one-recall: 0.54 one-ratio: 1.08127
iteration: 4 recall: 0.914 accuracy: 0.00760552 cost: 0.0249113 M: 11.725 delta: 0.566234 time: 21.4996 one-recall: 0.98 one-ratio: 1.00447
iteration: 5 recall: 0.9796 accuracy: 0.000839275 cost: 0.0376857 M: 17.4226 delta: 0.22458 time: 30.3735 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9884 accuracy: 0.000429563 cost: 0.0460114 M: 21.1537 delta: 0.134182 time: 36.0811 one-recall: 1 one-ratio: 1
iteration: 7 recall: 0.9896 accuracy: 0.000361085 cost: 0.0477748 M: 21.8102 delta: 0.126942 time: 37.4389 one-recall: 1 one-ratio: 1
iteration: 8 recall: 0.9904 accuracy: 0.000331267 cost: 0.0480915 M: 21.9263 delta: 0.125892 time: 37.7541 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 38.09999999999991
Index size:  103072.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0004450000
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0300000000, query time of that 0.0141345290, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2400000000, query time of that 0.1273771510, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 2.5000000000, query time of that 1.2823397550, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1400000000, query time of that 0.0135705770, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.3600000000, query time of that 0.1432249150, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Accept!
  -> Decision True in time 13.7500000000, query time of that 1.4368858800, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0161211930, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1275.69 < 1312.9
  -> Decision False in time 2.4400000000, query time of that 0.0302997500, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2030.91 < 2170.27
  -> Decision False in time 102.7000000000, query time of that 1.2048831950, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 30, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 30, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0048 accuracy: 1.73934 cost: 0.00633344 M: 10 delta: 1 time: 6.87667 one-recall: 0.01 one-ratio: 1.96915
iteration: 2 recall: 0.0692 accuracy: 0.571368 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.501 one-recall: 0.06 one-ratio: 1.45468
iteration: 3 recall: 0.4896 accuracy: 0.114628 cost: 0.0167507 M: 11.1153 delta: 0.845797 time: 15.5364 one-recall: 0.48 one-ratio: 1.13365
iteration: 4 recall: 0.936 accuracy: 0.0065462 cost: 0.0249116 M: 11.7248 delta: 0.566215 time: 21.4967 one-recall: 0.95 one-ratio: 1.00672
iteration: 5 recall: 0.9892 accuracy: 0.000607385 cost: 0.0376867 M: 17.4238 delta: 0.224554 time: 30.373 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9944 accuracy: 0.000248872 cost: 0.0460159 M: 21.1566 delta: 0.134129 time: 36.0848 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.42000000000007
Index size:  99528.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0010083333
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0068011270, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.1900000000, query time of that 0.0694915480, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 1.9100000000, query time of that 0.6945703110, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0074585430, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.2700000000, query time of that 0.0772761010, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Accept!
  -> Decision True in time 12.8700000000, query time of that 0.7757253510, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3600000000, query time of that 0.0093357810, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1802.07 < 1892.25
  -> Decision False in time 4.0200000000, query time of that 0.0283480470, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1755.09 < 1759.03
  -> Decision False in time 70.9400000000, query time of that 0.4749767900, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 5, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 5, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0072 accuracy: 1.51445 cost: 0.00633344 M: 10 delta: 1 time: 6.8788 one-recall: 0 one-ratio: 1.86417
iteration: 2 recall: 0.0708 accuracy: 0.513578 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5015 one-recall: 0.06 one-ratio: 1.31899
iteration: 3 recall: 0.482 accuracy: 0.114026 cost: 0.0167507 M: 11.1153 delta: 0.84578 time: 15.537 one-recall: 0.49 one-ratio: 1.08758
iteration: 4 recall: 0.9252 accuracy: 0.00831879 cost: 0.0249131 M: 11.7249 delta: 0.566216 time: 21.4979 one-recall: 0.96 one-ratio: 1.00245
iteration: 5 recall: 0.9912 accuracy: 0.000407839 cost: 0.0376754 M: 17.4182 delta: 0.224652 time: 30.3653 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 30.680000000000064
Index size:  92568.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0113833333
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0048810930, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Reject!
1712.42 < 2246.31
  -> Decision False in time 0.0800000000, query time of that 0.0203592040, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
2045.25 < 2079.01
  -> Decision False in time 0.0700000000, query time of that 0.0176861890, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Reject!
2524.27 < 3023.3
  -> Decision False in time 0.0100000000, query time of that 0.0006831910, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Reject!
1518.6 < 1546.89
  -> Decision False in time 1.1000000000, query time of that 0.0437516310, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
2109.38 < 2724.6
  -> Decision False in time 0.0500000000, query time of that 0.0023504060, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0059579540, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
2294.12 < 2483.31
  -> Decision False in time 5.5500000000, query time of that 0.0242373060, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2007 < 2033.45
  -> Decision False in time 0.6300000000, query time of that 0.0029985430, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 90, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 90, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.006 accuracy: 1.82424 cost: 0.00633344 M: 10 delta: 1 time: 6.87572 one-recall: 0.02 one-ratio: 1.91185
iteration: 2 recall: 0.076 accuracy: 0.586263 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.4998 one-recall: 0.07 one-ratio: 1.37943
iteration: 3 recall: 0.4732 accuracy: 0.129956 cost: 0.0167507 M: 11.1153 delta: 0.84581 time: 15.5373 one-recall: 0.49 one-ratio: 1.11703
iteration: 4 recall: 0.9204 accuracy: 0.00964779 cost: 0.0249133 M: 11.725 delta: 0.5662 time: 21.5022 one-recall: 0.96 one-ratio: 1.0048
iteration: 5 recall: 0.9892 accuracy: 0.000775498 cost: 0.0376843 M: 17.4206 delta: 0.224576 time: 30.3806 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9976 accuracy: 0.000155341 cost: 0.0460046 M: 21.1523 delta: 0.134229 time: 36.0898 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.42000000000007
Index size:  99516.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0004466667
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0300000000, query time of that 0.0138641140, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2400000000, query time of that 0.1225886840, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 2.4000000000, query time of that 1.1812439920, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1400000000, query time of that 0.0145571480, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.3600000000, query time of that 0.1334462420, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Accept!
  -> Decision True in time 13.7800000000, query time of that 1.3429993930, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3800000000, query time of that 0.0151322740, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Accept!
  -> Decision True in time 13.4400000000, query time of that 0.1475633460, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2388.74 < 2570.83
  -> Decision False in time 133.9900000000, query time of that 1.4527209740, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 40, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 40, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0028 accuracy: 1.64054 cost: 0.00633344 M: 10 delta: 1 time: 6.87946 one-recall: 0 one-ratio: 1.91755
iteration: 2 recall: 0.0704 accuracy: 0.572508 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5025 one-recall: 0.05 one-ratio: 1.38146
iteration: 3 recall: 0.4724 accuracy: 0.118556 cost: 0.0167507 M: 11.1153 delta: 0.845796 time: 15.5404 one-recall: 0.52 one-ratio: 1.08798
iteration: 4 recall: 0.9288 accuracy: 0.005925 cost: 0.0249116 M: 11.7249 delta: 0.566231 time: 21.5018 one-recall: 0.97 one-ratio: 1.00571
iteration: 5 recall: 0.99 accuracy: 0.00050381 cost: 0.0376856 M: 17.4235 delta: 0.224565 time: 30.3793 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9968 accuracy: 0.000100978 cost: 0.0460227 M: 21.1582 delta: 0.13411 time: 36.0978 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.440000000000055
Index size:  99520.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0030200000
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0093373670, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2000000000, query time of that 0.0813763750, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
2839.27 < 2843.15
  -> Decision False in time 0.1700000000, query time of that 0.0651023060, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0078104040, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Reject!
2486.23 < 2510.77
  -> Decision False in time 0.9400000000, query time of that 0.0669761970, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
2453.3 < 2465.58
  -> Decision False in time 7.1600000000, query time of that 0.4915426510, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0097599000, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Accept!
  -> Decision True in time 13.3600000000, query time of that 0.1021769820, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2151.79 < 2175.85
  -> Decision False in time 47.8400000000, query time of that 0.3675639550, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 60, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 60, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.006 accuracy: 1.3997 cost: 0.00633344 M: 10 delta: 1 time: 6.8892 one-recall: 0.01 one-ratio: 1.73277
iteration: 2 recall: 0.0716 accuracy: 0.471885 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.512 one-recall: 0.07 one-ratio: 1.32772
iteration: 3 recall: 0.4556 accuracy: 0.104181 cost: 0.0167507 M: 11.1153 delta: 0.845796 time: 15.5454 one-recall: 0.52 one-ratio: 1.07871
iteration: 4 recall: 0.8928 accuracy: 0.00866536 cost: 0.0249124 M: 11.725 delta: 0.566208 time: 21.5047 one-recall: 0.96 one-ratio: 1.00271
iteration: 5 recall: 0.9852 accuracy: 0.000582193 cost: 0.0376818 M: 17.4205 delta: 0.224595 time: 30.3753 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.992 accuracy: 0.000272659 cost: 0.0460199 M: 21.1567 delta: 0.134152 time: 36.0902 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.42000000000007
Index size:  99520.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0010166667
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0300000000, query time of that 0.0106423980, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2100000000, query time of that 0.0934556600, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Accept!
  -> Decision True in time 2.1700000000, query time of that 0.9430215180, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0110588580, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.3300000000, query time of that 0.1057288690, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
1584.09 < 1641.4
  -> Decision False in time 11.0900000000, query time of that 0.8868886080, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0117201970, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Accept!
  -> Decision True in time 13.4900000000, query time of that 0.1207703640, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2479.27 < 2552.57
  -> Decision False in time 3.4300000000, query time of that 0.0307832970, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 2, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 2, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0056 accuracy: 1.83732 cost: 0.00633344 M: 10 delta: 1 time: 6.88499 one-recall: 0.01 one-ratio: 1.96703
iteration: 2 recall: 0.0636 accuracy: 0.637156 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5074 one-recall: 0.06 one-ratio: 1.41779
iteration: 3 recall: 0.4644 accuracy: 0.131139 cost: 0.0167507 M: 11.1153 delta: 0.845803 time: 15.5433 one-recall: 0.47 one-ratio: 1.09979
iteration: 4 recall: 0.9156 accuracy: 0.00843653 cost: 0.0249102 M: 11.7247 delta: 0.566264 time: 21.5039 one-recall: 0.99 one-ratio: 1.00005
iteration: 5 recall: 0.988 accuracy: 0.000674411 cost: 0.0376872 M: 17.4231 delta: 0.224514 time: 30.3802 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9928 accuracy: 0.000473247 cost: 0.0460158 M: 21.1553 delta: 0.13413 time: 36.0912 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.430000000000064
Index size:  99516.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0151333333
  Testing...
|S| = 20
|T| = 283
Reject!
2070.41 < 2460.82
  -> Decision False in time 0.0000000000, query time of that 0.0002408360, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Reject!
1611.11 < 2282.5
  -> Decision False in time 0.0600000000, query time of that 0.0169380060, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
2353.81 < 2609.8
  -> Decision False in time 0.1400000000, query time of that 0.0363649180, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Reject!
1480.47 < 2157.23
  -> Decision False in time 0.0500000000, query time of that 0.0026859340, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.2400000000, query time of that 0.0508168310, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
1520.11 < 1531.68
  -> Decision False in time 0.4500000000, query time of that 0.0183102040, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0066774120, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1372.24 < 1388.5
  -> Decision False in time 4.9700000000, query time of that 0.0243415090, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
2007 < 2033.45
  -> Decision False in time 4.1500000000, query time of that 0.0199826150, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 10, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 10, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0092 accuracy: 1.78563 cost: 0.00633344 M: 10 delta: 1 time: 6.88222 one-recall: 0.02 one-ratio: 1.93856
iteration: 2 recall: 0.0672 accuracy: 0.627302 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5049 one-recall: 0.04 one-ratio: 1.4066
iteration: 3 recall: 0.444 accuracy: 0.13727 cost: 0.0167507 M: 11.1153 delta: 0.8458 time: 15.5407 one-recall: 0.49 one-ratio: 1.12713
iteration: 4 recall: 0.927999 accuracy: 0.00606872 cost: 0.0249113 M: 11.7249 delta: 0.566208 time: 21.5005 one-recall: 0.97 one-ratio: 1.00424
iteration: 5 recall: 0.988 accuracy: 0.000569729 cost: 0.03768 M: 17.4208 delta: 0.224625 time: 30.3704 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.994 accuracy: 0.000243263 cost: 0.0460118 M: 21.1555 delta: 0.134181 time: 36.081 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 36.42000000000007
Index size:  99520.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0027033333
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0058682110, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Reject!
1517.22 < 1541.03
  -> Decision False in time 0.0200000000, query time of that 0.0056249060, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
3780.72 < 3782.56
  -> Decision False in time 1.1100000000, query time of that 0.3216692150, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1400000000, query time of that 0.0060418830, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.2800000000, query time of that 0.0577868300, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
1550.09 < 1602.61
  -> Decision False in time 8.8500000000, query time of that 0.3971152510, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3900000000, query time of that 0.0066056920, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1156.64 < 1163.26
  -> Decision False in time 3.0200000000, query time of that 0.0167234560, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1610.69 < 1621.77
  -> Decision False in time 3.7400000000, query time of that 0.0182192200, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 70, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 70, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0088 accuracy: 1.71368 cost: 0.00633344 M: 10 delta: 1 time: 6.88563 one-recall: 0.02 one-ratio: 1.8364
iteration: 2 recall: 0.0728 accuracy: 0.566279 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5099 one-recall: 0.08 one-ratio: 1.37305
iteration: 3 recall: 0.4884 accuracy: 0.116356 cost: 0.0167507 M: 11.1153 delta: 0.845791 time: 15.5462 one-recall: 0.49 one-ratio: 1.10098
iteration: 4 recall: 0.9068 accuracy: 0.00908947 cost: 0.0249127 M: 11.7247 delta: 0.566224 time: 21.5063 one-recall: 0.95 one-ratio: 1.01062
iteration: 5 recall: 0.9792 accuracy: 0.00133202 cost: 0.0376844 M: 17.4224 delta: 0.224562 time: 30.3793 one-recall: 1 one-ratio: 1
iteration: 6 recall: 0.9896 accuracy: 0.000607145 cost: 0.0460175 M: 21.1563 delta: 0.134126 time: 36.0918 one-recall: 1 one-ratio: 1
iteration: 7 recall: 0.9908 accuracy: 0.000534631 cost: 0.0477879 M: 21.8145 delta: 0.126911 time: 37.4547 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 37.789999999999964
Index size:  102520.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0004333333
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0300000000, query time of that 0.0114109410, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Accept!
  -> Decision True in time 0.2200000000, query time of that 0.1062145330, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
1962.12 < 1970.41
  -> Decision False in time 0.6400000000, query time of that 0.2935986780, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Accept!
  -> Decision True in time 0.1300000000, query time of that 0.0117453240, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Accept!
  -> Decision True in time 1.3400000000, query time of that 0.1195212620, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Accept!
  -> Decision True in time 13.6100000000, query time of that 1.1895035380, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0142650500, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
1262.66 < 1377.48
  -> Decision False in time 4.7500000000, query time of that 0.0463169760, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1635.28 < 1649.62
  -> Decision False in time 75.5500000000, query time of that 0.7449196670, with c1=5.0000000000, c2=0.1000000000
Definition(algorithm='kgraph', constructor='KGraph', module='ann_benchmarks.algorithms.kgraph', docker_tag='ann-benchmarks-kgraph', arguments=['euclidean', 3, {'reverse': -1}, False]) ...
Trying to instantiate ann_benchmarks.algorithms.kgraph.KGraph(['euclidean', 3, {'reverse': -1}, False])
Got a train set of size (60000 * 784)
Generating control...
Initializing...
iteration: 1 recall: 0.0044 accuracy: 1.62202 cost: 0.00633344 M: 10 delta: 1 time: 6.88433 one-recall: 0 one-ratio: 1.88164
iteration: 2 recall: 0.0676 accuracy: 0.531483 cost: 0.0102345 M: 10 delta: 0.893354 time: 10.5069 one-recall: 0.05 one-ratio: 1.39538
iteration: 3 recall: 0.4676 accuracy: 0.111829 cost: 0.0167507 M: 11.1153 delta: 0.845794 time: 15.5411 one-recall: 0.53 one-ratio: 1.10903
iteration: 4 recall: 0.9268 accuracy: 0.00666177 cost: 0.0249111 M: 11.7245 delta: 0.566233 time: 21.5008 one-recall: 0.95 one-ratio: 1.01103
iteration: 5 recall: 0.9924 accuracy: 0.000416861 cost: 0.0376834 M: 17.4231 delta: 0.224527 time: 30.3737 one-recall: 1 one-ratio: 1
Graph completion with reverse edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Reranking edges...

0%   10   20   30   40   50   60   70   80   90   100%
|----|----|----|----|----|----|----|----|----|----|
***************************************************
Built index in 30.679999999999836
Index size:  92572.0
Run 1/1...
  Calculating distance...
  -> Distance: 0.0124583333
  Testing...
|S| = 20
|T| = 283
Accept!
  -> Decision True in time 0.0200000000, query time of that 0.0045979100, with c1=0.0500000000, c2=0.0010000000
|S| = 196
|T| = 283
Reject!
1174.24 < 1179.17
  -> Decision False in time 0.0300000000, query time of that 0.0069196260, with c1=0.0500000000, c2=0.0100000000
|S| = 1960
|T| = 283
Reject!
2578.03 < 2892.06
  -> Decision False in time 0.0200000000, query time of that 0.0053990070, with c1=0.0500000000, c2=0.1000000000
|S| = 20
|T| = 2821
Reject!
1395.61 < 1404.29
  -> Decision False in time 0.0800000000, query time of that 0.0031582750, with c1=0.5000000000, c2=0.0010000000
|S| = 196
|T| = 2821
Reject!
2134.37 < 2267.3
  -> Decision False in time 0.1400000000, query time of that 0.0052880470, with c1=0.5000000000, c2=0.0100000000
|S| = 1960
|T| = 2821
Reject!
2287.38 < 3024.02
  -> Decision False in time 0.1900000000, query time of that 0.0081964750, with c1=0.5000000000, c2=0.1000000000
|S| = 20
|T| = 28201
Accept!
  -> Decision True in time 1.3700000000, query time of that 0.0060809060, with c1=5.0000000000, c2=0.0010000000
|S| = 196
|T| = 28201
Reject!
2004.37 < 2095.12
  -> Decision False in time 0.6200000000, query time of that 0.0029340390, with c1=5.0000000000, c2=0.0100000000
|S| = 1960
|T| = 28201
Reject!
1662.1 < 1871.18
  -> Decision False in time 13.5600000000, query time of that 0.0595607280, with c1=5.0000000000, c2=0.1000000000
