ML之KMeans：利用KMeans算法对Boston房价数据集(两特征+归一化)进行二聚类分析

飞机
飞机 2022-09-19 11:54:20 49167
分类专栏： 资讯

ML之KMeans：利用KMeans算法对Boston房价数据集(两特征+归一化)进行二聚类分析

目录

利用KMeans算法对Boston房价数据集(两特征+归一化)进行二聚类分析

设计思路

输出结果

核心代码

相关文章
ML之KMeans：利用KMeans算法对Boston房价数据集(两特征+归一化)进行二聚类分析
ML之KMeans：利用KMeans算法对Boston房价数据集(两特征+归一化)进行二聚类分析实现
 

利用KMeans算法对Boston房价数据集(两特征+归一化)进行二聚类分析

设计思路

输出结果

  1. train_boston_data.shape (1460, 81)
  2.    Id  MSSubClass MSZoning  ...  SaleType  SaleCondition SalePrice
  3. 0   1          60       RL  ...        WD         Normal    208500
  4. 1   2          20       RL  ...        WD         Normal    181500
  5. 2   3          60       RL  ...        WD         Normal    223500
  6. 3   4          70       RL  ...        WD        Abnorml    140000
  7. 4   5          60       RL  ...        WD         Normal    250000
  8.  
  9. [5 rows x 81 columns]
  10. train_t.head()    LotFrontage  GarageArea  SalePrice
  11. 0         65.0         548     208500
  12. 1         80.0         460     181500
  13. 2         68.0         608     223500
  14. 3         60.0         642     140000
  15. 4         84.0         836     250000
  16. after scale,train_t.head()    LotFrontage  GarageArea  SalePrice
  17. 0     0.207668    0.386460   0.276159
  18. 1     0.255591    0.324401   0.240397
  19. 2     0.217252    0.428773   0.296026
  20. 3     0.191693    0.452750   0.185430
  21. 4     0.268371    0.589563   0.331126
  22.    LotFrontage  GarageArea
  23. 0     0.207668    0.386460
  24. 1     0.255591    0.324401
  25. 2     0.217252    0.428773
  26. 3     0.191693    0.452750
  27. 4     0.268371    0.589563
  28.                      Id  MSSubClass  LotFrontage  ...    MoSold    YrSold  SalePrice
  29. Id             1.000000    0.011156    -0.010601  ...  0.021172  0.000712  -0.021917
  30. MSSubClass     0.011156    1.000000    -0.386347  ... -0.013585 -0.021407  -0.084284
  31. LotFrontage   -0.010601   -0.386347     1.000000  ...  0.011200  0.007450   0.351799
  32. LotArea       -0.033226   -0.139781     0.426095  ...  0.001205 -0.014261   0.263843
  33. OverallQual   -0.028365    0.032628     0.251646  ...  0.070815 -0.027347   0.790982
  34. OverallCond    0.012609   -0.059316    -0.059213  ... -0.003511  0.043950  -0.077856
  35. YearBuilt     -0.012713    0.027850     0.123349  ...  0.012398 -0.013618   0.522897
  36. YearRemodAdd  -0.021998    0.040581     0.088866  ...  0.021490  0.035743   0.507101
  37. MasVnrArea    -0.050298    0.022936     0.193458  ... -0.005965 -0.008201   0.477493
  38. BsmtFinSF1    -0.005024   -0.069836     0.233633  ... -0.015727  0.014359   0.386420
  39. BsmtFinSF2    -0.005968   -0.065649     0.049900  ... -0.015211  0.031706  -0.011378
  40. BsmtUnfSF     -0.007940   -0.140759     0.132644  ...  0.034888 -0.041258   0.214479
  41. TotalBsmtSF   -0.015415   -0.238518     0.392075  ...  0.013196 -0.014969   0.613581
  42. 1stFlrSF       0.010496   -0.251758     0.457181  ...  0.031372 -0.013604   0.605852
  43. 2ndFlrSF       0.005590    0.307886     0.080177  ...  0.035164 -0.028700   0.319334
  44. LowQualFinSF  -0.044230    0.046474     0.038469  ... -0.022174 -0.028921  -0.025606
  45. GrLivArea      0.008273    0.074853     0.402797  ...  0.050240 -0.036526   0.708624
  46. BsmtFullBath   0.002289    0.003491     0.100949  ... -0.025361  0.067049   0.227122
  47. BsmtHalfBath  -0.020155   -0.002333    -0.007234  ...  0.032873 -0.046524  -0.016844
  48. FullBath       0.005587    0.131608     0.198769  ...  0.055872 -0.019669   0.560664
  49. HalfBath       0.006784    0.177354     0.053532  ... -0.009050 -0.010269   0.284108
  50. BedroomAbvGr   0.037719   -0.023438     0.263170  ...  0.046544 -0.036014   0.168213
  51. KitchenAbvGr   0.002951    0.281721    -0.006069  ...  0.026589  0.031687  -0.135907
  52. TotRmsAbvGrd   0.027239    0.040380     0.352096  ...  0.036907 -0.034516   0.533723
  53. Fireplaces    -0.019772   -0.045569     0.266639  ...  0.046357 -0.024096   0.466929
  54. GarageYrBlt    0.000072    0.085072     0.070250  ...  0.005337 -0.001014   0.486362
  55. GarageCars     0.016570   -0.040110     0.285691  ...  0.040522 -0.039117   0.640409
  56. GarageArea     0.017634   -0.098672     0.344997  ...  0.027974 -0.027378   0.623431
  57. WoodDeckSF    -0.029643   -0.012579     0.088521  ...  0.021011  0.022270   0.324413
  58. OpenPorchSF   -0.000477   -0.006100     0.151972  ...  0.071255 -0.057619   0.315856
  59. EnclosedPorch  0.002889   -0.012037     0.010700  ... -0.028887 -0.009916  -0.128578
  60. 3SsnPorch     -0.046635   -0.043825     0.070029  ...  0.029474  0.018645   0.044584
  61. ScreenPorch    0.001330   -0.026030     0.041383  ...  0.023217  0.010694   0.111447
  62. PoolArea       0.057044    0.008283     0.206167  ... -0.033737 -0.059689   0.092404
  63. MiscVal       -0.006242   -0.007683     0.003368  ... -0.006495  0.004906  -0.021190
  64. MoSold         0.021172   -0.013585     0.011200  ...  1.000000 -0.145721   0.046432
  65. YrSold         0.000712   -0.021407     0.007450  ... -0.145721  1.000000  -0.028923
  66. SalePrice     -0.021917   -0.084284     0.351799  ...  0.046432 -0.028923   1.000000
  67.  
  68. [38 rows x 38 columns]
  69. k_means_cluster_centers 
  70.  [[0.1938454  0.21080405]
  71.  [0.25140958 0.44595543]]
  72. k_means_labels_unique 
  73.  [0 1]
  74. 0 [1 1 1 ... 0 0 0]
  75. 0 [1 1 1 ... 0 0 0] [False False False ...  True  True  True]
  76. 1 [1 1 1 ... 0 0 0]
  77. 1 [1 1 1 ... 0 0 0] [ True  True  True ... False False False]

核心代码

  1. class KMeans Found at: sklearn.cluster._kmeans
  2.  
  3. class KMeans(TransformerMixin, ClusterMixin, BaseEstimator):
  4.     """K-Means clustering.
  5.     
  6.     Read more in the :ref:`User Guide <k_means>`.
  7.     
  8.     Parameters
  9.     ----------
  10.     
  11.     n_clusters : int, default=8
  12.     The number of clusters to form as well as the number of
  13.     centroids to generate.
  14.     
  15.     init : {'k-means++', 'random', ndarray, callable}, default='k-
  16.      means++'
  17.     Method for initialization:
  18.     
  19.     'k-means++' : selects initial cluster centers for k-mean
  20.     clustering in a smart way to speed up convergence. See 
  21.      section
  22.     Notes in k_init for more details.
  23.     
  24.     'random': choose `n_clusters` observations (rows) at 
  25.      random from data
  26.     for the initial centroids.
  27.     
  28.     If an ndarray is passed, it should be of shape (n_clusters, 
  29.      n_features)
  30.     and gives the initial centers.
  31.     
  32.     If a callable is passed, it should take arguments X, 
  33.      n_clusters and a
  34.     random state and return an initialization.
  35.     
  36.     n_init : int, default=10
  37.     Number of time the k-means algorithm will be run with 
  38.      different
  39.     centroid seeds. The final results will be the best output of
  40.     n_init consecutive runs in terms of inertia.
  41.     
  42.     max_iter : int, default=300
  43.     Maximum number of iterations of the k-means algorithm 
  44.      for a
  45.     single run.
  46.     
  47.     tol : float, default=1e-4
  48.     Relative tolerance with regards to Frobenius norm of the 
  49.      difference
  50.     in the cluster centers of two consecutive iterations to 
  51.      declare
  52.     convergence.
  53.     It's not advised to set `tol=0` since convergence might 
  54.      never be
  55.     declared due to rounding errors. Use a very small number 
  56.      instead.
  57.     
  58.     precompute_distances : {'auto', True, False}, default='auto'
  59.     Precompute distances (faster but takes more memory).
  60.     
  61.     'auto' : do not precompute distances if n_samples * 
  62.      n_clusters > 12
  63.     million. This corresponds to about 100MB overhead per 
  64.      job using
  65.     double precision.
  66.     
  67.     True : always precompute distances.
  68.     
  69.     False : never precompute distances.
  70.     
  71.     .. deprecated:: 0.23
  72.     'precompute_distances' was deprecated in version 0.22 
  73.      and will be
  74.     removed in 0.25. It has no effect.
  75.     
  76.     verbose : int, default=0
  77.     Verbosity mode.
  78.     
  79.     random_state : int, RandomState instance, default=None
  80.     Determines random number generation for centroid 
  81.      initialization. Use
  82.     an int to make the randomness deterministic.
  83.     See :term:`Glossary <random_state>`.
  84.     
  85.     copy_x : bool, default=True
  86.     When pre-computing distances it is more numerically 
  87.      accurate to center
  88.     the data first. If copy_x is True (default), then the original 
  89.      data is
  90.     not modified. If False, the original data is modified, and put 
  91.      back
  92.     before the function returns, but small numerical 
  93.      differences may be
  94.     introduced by subtracting and then adding the data mean. 
  95.      Note that if
  96.     the original data is not C-contiguous, a copy will be made 
  97.      even if
  98.     copy_x is False. If the original data is sparse, but not in CSR 
  99.      format,
  100.     a copy will be made even if copy_x is False.
  101.     
  102.     n_jobs : int, default=None
  103.     The number of OpenMP threads to use for the 
  104.      computation. Parallelism is
  105.     sample-wise on the main cython loop which assigns each 
  106.      sample to its
  107.     closest center.
  108.     
  109.     ``None`` or ``-1`` means using all processors.
  110.     
  111.     .. deprecated:: 0.23
  112.     ``n_jobs`` was deprecated in version 0.23 and will be 
  113.      removed in
  114.     0.25.
  115.     
  116.     algorithm : {"auto", "full", "elkan"}, default="auto"
  117.     K-means algorithm to use. The classical EM-style algorithm 
  118.      is "full".
  119.     The "elkan" variation is more efficient on data with well-
  120.      defined
  121.     clusters, by using the triangle inequality. However it's 
  122.      more memory
  123.     intensive due to the allocation of an extra array of shape
  124.     (n_samples, n_clusters).
  125.     
  126.     For now "auto" (kept for backward compatibiliy) chooses 
  127.      "elkan" but it
  128.     might change in the future for a better heuristic.
  129.     
  130.     .. versionchanged:: 0.18
  131.     Added Elkan algorithm
  132.     
  133.     Attributes
  134.     ----------
  135.     cluster_centers_ : ndarray of shape (n_clusters, n_features)
  136.     Coordinates of cluster centers. If the algorithm stops 
  137.      before fully
  138.     converging (see ``tol`` and ``max_iter``), these will not be
  139.     consistent with ``labels_``.
  140.     
  141.     labels_ : ndarray of shape (n_samples,)
  142. 
            
    网站声明：如果转载，请联系本站管理员。否则一切后果自行承担。
    
          
    
        
          
    
            
              赞同  0
            
            
          
    
      
    
      
      
    
        评论 0 条
        
      
		              
    
		    
    
        
              
    
      
      
    
        
    相关文章
飞机
飞机L0
粉丝 0 发表 9 + 关注 私信
上周热门
如何使用 StarRocks 管理和优化数据湖中的数据？  2941
【软件正版化】软件正版化工作要点  2860
统信UOS试玩黑神话：悟空  2819
信刻光盘安全隔离与信息交换系统  2712
镜舟科技与中启乘数科技达成战略合作，共筑数据服务新生态  1246
grub引导程序无法找到指定设备和分区  1213
华为全联接大会2024丨软通动力分论坛精彩议程抢先看！  163
点击报名 | 京东2025校招进校行程预告  162
2024海洋能源产业融合发展论坛暨博览会同期活动-海洋能源与数字化智能化论坛成功举办  160
华为纯血鸿蒙正式版9月底见！但Mate 70的内情还得接着挖...  157
本周热议
我的信创开放社区兼职赚钱历程 40
今天你签到了吗？ 27
信创开放社区邀请他人注册的具体步骤如下 15
如何玩转信创开放社区—从小白进阶到专家 15
方德桌面操作系统 14
我有15积分有什么用? 13
用抖音玩法闯信创开放社区——用平台宣传企业产品服务 13
如何让你先人一步获得悬赏问题信息？（创作者必看） 12
2024中国信创产业发展大会暨中国信息科技创新与应用博览会 9
中央国家机关政府采购中心：应当将CPU、操作系统符合安全可靠测评要求纳入采购需求 8
热门标签更多

  • 关注我们

    信创开放社区公众号二维码

    扫一扫关注公众号

    信创开放社区交流群

    扫一扫加入交流群

加入交流群

请使用微信扫一扫!