test0.save

高效的k-means算法实现

------------------------------------------------------------
kmltest: KMlocal (k-means clustering by local search)
    Version: 1.7 (Use at your own risk)
    Copyright: David M. Mount.
    Latest Revision: August 10, 2005.
------------------------------------------------------------

stats = tree
[Read Data Points:
  data_size  = 20
  file_name  = test1-dat.txt
  dim        = 2
]
Contents of the kc-tree: [
    ....Leaf n=1 <13> sm=[     0.91     0.87 ] ss=1.585
    ...Split cd=0 cv=0.4475 nd=4 sm=[     1.19     3.56 ] ss=4.027
    ......Leaf n=1 <17> sm=[     0.13     0.92 ] ss=0.8633
    .....Split cd=0 cv=  0.13 nd=2 sm=[     0.18     1.81 ] ss=1.658
    ......Leaf n=1 <19> sm=[     0.05     0.89 ] ss=0.7946
    ....Split cd=1 cv=  0.88 nd=3 sm=[     0.28     2.69 ] ss=2.442
    .....Leaf n=1 <6> sm=[      0.1     0.88 ] ss=0.7844
    ..Split cd=1 cv=   0.5 nd=7 sm=[     3.01     4.23 ] ss=5.695
    .....Leaf n=1 <3> sm=[     0.92     0.47 ] ss=1.067
    ....Split cd=1 cv=  0.26 nd=2 sm=[     1.66     0.63 ] ss=1.64
    .....Leaf n=1 <7> sm=[     0.74     0.16 ] ss=0.5732
    ...Split cd=0 cv=0.4475 nd=3 sm=[     1.82     0.67 ] ss=1.668
    ....Leaf n=1 <4> sm=[     0.16     0.04 ] ss=0.0272
    .Split cd=0 cv=-0.025 nd=11 sm=[     1.42     6.26 ] ss=7.984
    ....Leaf n=1 <18> sm=[    -0.08     0.61 ] ss=0.3785
    ...Split cd=0 cv=-0.4975 nd=2 sm=[    -0.59     1.59 ] ss=1.599
    ....Leaf n=1 <16> sm=[    -0.51     0.98 ] ss=1.22
    ..Split cd=1 cv=   0.5 nd=4 sm=[    -1.59     2.03 ] ss=2.289
    ....Leaf n=1 <0> sm=[    -0.29     0.17 ] ss=0.113
    ...Split cd=0 cv=-0.4975 nd=2 sm=[       -1     0.44 ] ss=0.69
    ....Leaf n=1 <15> sm=[    -0.71     0.27 ] ss=0.577
    Split cd=1 cv=  0.02 nd=20 sm=[     4.72     2.09 ] ss=14.56
    .....Leaf n=1 <2> sm=[      0.9    -0.18 ] ss=0.8424
    ....Split cd=1 cv= -0.22 nd=3 sm=[     2.38    -0.92 ] ss=2.276
    ......Leaf n=1 <14> sm=[     0.92    -0.38 ] ss=0.9908
    .....Split cd=0 cv=0.6838 nd=2 sm=[     1.48    -0.74 ] ss=1.434
    ......Leaf n=1 <1> sm=[     0.56    -0.36 ] ss=0.4432
    ...Split cd=0 cv=0.4475 nd=4 sm=[     2.68       -1 ] ss=2.373
    ....Leaf n=1 <5> sm=[      0.3    -0.08 ] ss=0.0964
    ..Split cd=1 cv= -0.46 nd=8 sm=[     4.27    -3.95 ] ss=5.586
    .....Leaf n=1 <9> sm=[      0.6    -0.55 ] ss=0.6625
    ....Split cd=1 cv=  -0.7 nd=2 sm=[     1.34    -1.49 ] ss=2.094
    .....Leaf n=1 <11> sm=[     0.74    -0.94 ] ss=1.431
    ...Split cd=0 cv=0.4475 nd=4 sm=[     1.59    -2.95 ] ss=3.213
    .....Leaf n=1 <10> sm=[     0.18    -0.64 ] ss=0.442
    ....Split cd=1 cv=  -0.7 nd=2 sm=[     0.25    -1.46 ] ss=1.119
    .....Leaf n=1 <8> sm=[     0.07    -0.82 ] ss=0.6773
    .Split cd=0 cv=-0.025 nd=9 sm=[      3.3    -4.17 ] ss=6.575
    ..Leaf n=1 <12> sm=[    -0.97    -0.22 ] ss=0.9893
]
       0	[        0        0 ] dist = 7.891e-309
       1	[        0        0 ] dist = 6.963e-77
       2	[        0        0 ] dist = 6.555e-260
       3	[        0        0 ] dist = 6.014e-154

[Run_k-means:
  k-means_alg      = lloyd
  data_size        = 20
  kcenters         = 4
  dim              = 2
  max_tot_stage    = 20
  max_run_stage    = 3
  min_accum_rdl    = 0.2
]
       0	[     0.18    -0.64 ] dist = 7.891e-309
       1	[     0.05     0.89 ] dist = 6.963e-77
       2	[    -0.71     0.27 ] dist = 6.555e-260
       3	[     0.56    -0.36 ] dist = 6.014e-154
	<stage: 0 curr: 0.1989 best: 0.1989 accumRDL: 0% >
       0	[    0.125    -0.73 ] dist =   0.0445
       1	[      0.1   0.8583 ] dist =    1.167
       2	[  -0.6567  0.07333 ] dist =   0.4941
       3	[   0.6489  -0.2022 ] dist =    2.272
	<stage: 1 curr: 0.1692 best: 0.1692 accumRDL: 14.92% >
       0	[     0.33     -0.8 ] dist =   0.4446
       1	[      0.1   0.8583 ] dist =    1.146
       2	[  -0.6567  0.07333 ] dist =   0.3695
       3	[   0.6375    -0.11 ] dist =    1.424
	<stage: 2 curr: 0.1557 best: 0.1557 accumRDL: 21.69% >
    <Generating new random centers>
       0	[   0.3975  -0.7375 ] dist =   0.4392
       1	[      0.1   0.8583 ] dist =    1.146
       2	[  -0.6567  0.07333 ] dist =   0.3695
       3	[   0.6429 -0.04714 ] dist =     1.16
	<stage: 3 curr: 0.1526 best: 0.1526 accumRDL: 1.981% >
       0	[   0.3975  -0.7375 ] dist =   0.4054
       1	[      0.1   0.8583 ] dist =    1.146
       2	[  -0.6567  0.07333 ] dist =   0.3695
       3	[   0.6429 -0.04714 ] dist =    1.132
	<stage: 4 curr: 0.1526 best: 0.1526 accumRDL: 1.981% >
       0	[   0.3975  -0.7375 ] dist =   0.4054
       1	[      0.1   0.8583 ] dist =    1.146
       2	[  -0.6567  0.07333 ] dist =   0.3695
       3	[   0.6429 -0.04714 ] dist =    1.132
	<stage: 5 curr: 0.1526 best: 0.1526 accumRDL: 1.981% >
    <Generating new random centers>
       0	[     0.91     0.87 ] dist =   0.4054
       1	[    -0.08     0.61 ] dist =    1.146
       2	[      0.1     0.88 ] dist =   0.3695
       3	[    -0.97    -0.22 ] dist =    1.132
	<stage: 6 curr: 0.7204 best: 0.1526 accumRDL: -362.6% >
       0	[    0.878    0.188 ] dist =    3.358
       1	[    0.185 -0.01625 ] dist =    7.796
       2	[  0.09333   0.8967 ] dist =   0.0051
       3	[  -0.3575  -0.3525 ] dist =    3.248
	<stage: 7 curr: 0.2675 best: 0.1526 accumRDL: 62.86% >
    <Generating new random centers>
       0	[    0.878    0.188 ] dist =    1.028
       1	[    0.345  -0.2867 ] dist =    2.159
       2	[   -0.062    0.856 ] dist =   0.4873
       3	[  -0.3575  -0.3525 ] dist =    1.677
	<stage: 8 curr: 0.2017 best: 0.1526 accumRDL: 24.6% >
    <Generating new random centers>
       0	[    0.878    0.188 ] dist =    1.028
       1	[   0.3729  -0.4786 ] dist =    1.467
       2	[   -0.062    0.856 ] dist =   0.3584
       3	[  -0.6567  0.07333 ] dist =    1.182
	<stage: 9 curr: 0.1472 best: 0.1472 accumRDL: 27.04% >
    <Generating new random centers>
       0	[   0.8675     0.33 ] dist =   0.7032
       1	[   0.4413  -0.4662 ] dist =    1.513
       2	[   -0.062    0.856 ] dist =   0.3584
       3	[  -0.6567  0.07333 ] dist =   0.3695
	<stage: 10 curr: 0.1412 best: 0.1412 accumRDL: 4.068% >
       0	[   0.8675     0.33 ] dist =   0.6221
       1	[   0.4413  -0.4662 ] dist =    1.474
       2	[   -0.062    0.856 ] dist =   0.3584
       3	[  -0.6567  0.07333 ] dist =   0.3695
	<stage: 11 curr: 0.1412 best: 0.1412 accumRDL: 4.068% >
       0	[   0.8675     0.33 ] dist =   0.6221
       1	[   0.4413  -0.4662 ] dist =    1.474
       2	[   -0.062    0.856 ] dist =   0.3584
       3	[  -0.6567  0.07333 ] dist =   0.3695
	<stage: 12 curr: 0.1412 best: 0.1412 accumRDL: 4.068% >
    <Generating new random centers>
       0	[      0.9    -0.18 ] dist =   0.6221
       1	[     0.92    -0.38 ] dist =    1.474
       2	[     0.16     0.04 ] dist =   0.3584
       3	[    -0.51     0.98 ] dist =   0.3695
	<stage: 13 curr: 0.3533 best: 0.1412 accumRDL: -140% >
       0	[   0.8675     0.33 ] dist =    1.667
       1	[    0.705  -0.5575 ] dist =   0.6073
       2	[ -0.09167  -0.2583 ] dist =    2.808
       3	[    -0.17   0.7583 ] dist =    1.983
	<stage: 14 curr: 0.1863 best: 0.1412 accumRDL: 47.25% >
    <Generating new random centers>
       0	[   0.8567      0.5 ] dist =   0.3609
       1	[    0.744   -0.482 ] dist =   0.4769
       2	[ -0.09167  -0.2583 ] dist =    1.894
       3	[    -0.17   0.7583 ] dist =   0.9945
	<stage: 15 curr: 0.1802 best: 0.1412 accumRDL: 3.305% >
       0	[   0.8567      0.5 ] dist =   0.2739
       1	[    0.744   -0.482 ] dist =   0.4408
       2	[ -0.09167  -0.2583 ] dist =    1.894
       3	[    -0.17   0.7583 ] dist =   0.9945
	<stage: 16 curr: 0.1802 best: 0.1412 accumRDL: 3.305% >
       0	[   0.8567      0.5 ] dist =   0.2739
       1	[    0.744   -0.482 ] dist =   0.4408
       2	[ -0.09167  -0.2583 ] dist =    1.894
       3	[    -0.17   0.7583 ] dist =   0.9945
	<stage: 17 curr: 0.1802 best: 0.1412 accumRDL: 3.305% >
    <Generating new random centers>
       0	[     0.91     0.87 ] dist =   0.2739
       1	[      0.9    -0.18 ] dist =   0.4408
       2	[    -0.97    -0.22 ] dist =    1.894
       3	[     0.07    -0.82 ] dist =   0.9945
	<stage: 18 curr: 0.391 best: 0.1412 accumRDL: -109.8% >
       0	[   0.3383   0.7733 ] dist =    3.215
       1	[   0.5971  -0.1929 ] dist =    1.523
       2	[    -0.62      0.3 ] dist =    2.574
       3	[     0.33     -0.8 ] dist =   0.5078
	<stage: 19 curr: 0.1692 best: 0.1412 accumRDL: 56.73% >
    <Generating new random centers>
       0	[   0.3383   0.7733 ] dist =    1.198
       1	[   0.5971  -0.1929 ] dist =   0.8793
       2	[    -0.62      0.3 ] dist =    1.002
       3	[     0.33     -0.8 ] dist =   0.3038
	<stage: 20 curr: 0.1692 best: 0.1412 accumRDL: 0% >
    <Generating new random centers>

[k-means completed:
  n_stages      = 20
  total_time    = 0.02 sec
  init_time     = 0 sec
  stage_time    = 0.001 sec/stage_(excl_init) 0.001 sec/stage_(incl_init)
  average_distort = 0.1412
  (Final Center Points:
       0	[   0.8675     0.33 ] dist =   0.6221
       1	[   0.4413  -0.4662 ] dist =    1.474
       2	[   -0.062    0.856 ] dist =   0.3584
       3	[  -0.6567  0.07333 ] dist =   0.3695
  )
]
  (Cluster assignments:
    Point  Center  Squared Dist
    -----  ------  ------------
       0       3       0.1438
       1       1      0.02539
       2       0       0.2612
       3       0      0.02236
       4       1       0.3354
       5       1       0.1691
       6       2      0.02682
       7       0      0.04516
       8       1        0.263
       9       1      0.03222
      10       1      0.09844
      11       1       0.3137
      12       3       0.1842
      13       0       0.2934
      14       1       0.2366
      15       3      0.04152
      16       2       0.2161
      17       2      0.04096
      18       2      0.06084
      19       2       0.0137
  )
  (Validating assignments.  Found 0 mismatches.)
<END_OF_RUN>