notebook.community

Edit and run



In [22]:

    
import pandas as pd
from sklearn import tree
from sklearn import metrics
from mlxtend.feature_selection import SequentialFeatureSelector
from sklearn.cross_validation import KFold, StratifiedKFold, cross_val_score
from sklearn.metrics import mean_squared_error, r2_score
from sklearn.pipeline import Pipeline
from sklearn.model_selection import cross_val_predict
from sklearn import linear_model
from mlxtend.feature_selection import SequentialFeatureSelector as SFS
from mlxtend.plotting import plot_sequential_feature_selection as plot_sfs
import matplotlib.pyplot as plt
import numpy as np
from __future__ import print_function



In [23]:

    
df_match = pd.read_csv('dataset.csv')
all_match_attributes=df_match.columns.values.tolist()
after_match_attributes =['home_team_goal', 'away_team_goal', 'possession_home', 'shoton_home', 'shoton_away', 
                      'shotoff_home', 'shotoff_away', 'corner_home', 'corner_away', 'cross_home', 'cross_away', 
                      'foulcommit_home', 'foulcommit_away', 'rcard_home', 'rcard_away', 'ycard_home', 'ycard_away', 
                      'throwin_home', 'throwin_away']
before_match_attributes = list(set(all_match_attributes) - set(after_match_attributes))
before_match_all=before_match_attributes+['possession_home']
df_before_match = df_match[before_match_all]



In [24]:

    
holdout_per=0.2
df_before_match_out_sample = df_before_match.sample(frac=holdout_per)
df_before_match_out_testtarget = df_before_match_out_sample['possession_home']
df_before_match_out_test =df_before_match_out_sample[before_match_attributes]

df_before_match_out_left = df_before_match.iloc[list(set(df_before_match.index.values.tolist())-set(df_before_match_out_sample.index.values.tolist()))]
df_before_match_out_training = df_before_match_out_left[before_match_attributes]
df_before_match_out_trainingtarget = df_before_match_out_left['possession_home']



In [25]:

    
lr = linear_model.LinearRegression()
sfs_greedy = SFS(lr, 
           k_features=df_before_match_out_training.shape[1], 
           forward=True, 
           floating=False, 
           verbose=2,
           scoring='neg_mean_squared_error',
           cv=10)
sfs_greedy = sfs_greedy.fit(df_before_match_out_training.as_matrix(), df_before_match_out_trainingtarget.as_matrix())
fig = plot_sfs(sfs_greedy.get_metric_dict(), kind='std_err')
fig.set_size_inches(18.5, 10.5, forward=True)
plt.title('Sequential Forward Selection (w. StdErr)')
plt.grid()
plt.show()









    



[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  41 out of  41 | elapsed:    0.4s finished

[2017-11-13 00:47:36] Features: 1/41 -- score: -58.7289698381[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  40 out of  40 | elapsed:    0.4s finished

[2017-11-13 00:47:36] Features: 2/41 -- score: -53.3892854073[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  39 out of  39 | elapsed:    0.6s finished

[2017-11-13 00:47:37] Features: 3/41 -- score: -52.5486830747[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  38 out of  38 | elapsed:    0.6s finished

[2017-11-13 00:47:37] Features: 4/41 -- score: -51.6461824979[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  37 out of  37 | elapsed:    0.5s finished

[2017-11-13 00:47:38] Features: 5/41 -- score: -51.2902865218[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  36 out of  36 | elapsed:    0.5s finished

[2017-11-13 00:47:38] Features: 6/41 -- score: -51.1051971651[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  35 out of  35 | elapsed:    0.6s finished

[2017-11-13 00:47:39] Features: 7/41 -- score: -50.8920514769[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  34 out of  34 | elapsed:    0.7s finished

[2017-11-13 00:47:40] Features: 8/41 -- score: -50.8261008571[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  33 out of  33 | elapsed:    0.6s finished

[2017-11-13 00:47:40] Features: 9/41 -- score: -50.7505603816[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  32 out of  32 | elapsed:    1.1s finished

[2017-11-13 00:47:41] Features: 10/41 -- score: -50.7005184699[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  31 out of  31 | elapsed:    0.6s finished

[2017-11-13 00:47:42] Features: 11/41 -- score: -50.6448873888[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  30 out of  30 | elapsed:    1.0s finished

[2017-11-13 00:47:43] Features: 12/41 -- score: -50.6056971801[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  29 out of  29 | elapsed:    0.8s finished

[2017-11-13 00:47:44] Features: 13/41 -- score: -50.5971663817[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  28 out of  28 | elapsed:    0.7s finished

[2017-11-13 00:47:45] Features: 14/41 -- score: -50.5178569556[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:    0.7s finished

[2017-11-13 00:47:45] Features: 15/41 -- score: -50.5094775166[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  26 out of  26 | elapsed:    0.7s finished

[2017-11-13 00:47:46] Features: 16/41 -- score: -50.51439094[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:    0.9s finished

[2017-11-13 00:47:47] Features: 17/41 -- score: -50.5348768024[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  24 out of  24 | elapsed:    0.6s finished

[2017-11-13 00:47:48] Features: 18/41 -- score: -50.5233740871[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  23 out of  23 | elapsed:    0.6s finished

[2017-11-13 00:47:48] Features: 19/41 -- score: -50.5402787234[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  22 out of  22 | elapsed:    0.6s finished

[2017-11-13 00:47:49] Features: 20/41 -- score: -50.5578008215[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  21 out of  21 | elapsed:    0.6s finished

[2017-11-13 00:47:49] Features: 21/41 -- score: -50.5815246656[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  20 out of  20 | elapsed:    0.6s finished

[2017-11-13 00:47:50] Features: 22/41 -- score: -50.5886990181[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  19 out of  19 | elapsed:    0.6s finished

[2017-11-13 00:47:51] Features: 23/41 -- score: -50.6243855568[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:    0.6s finished

[2017-11-13 00:47:51] Features: 24/41 -- score: -50.6708489597[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  17 out of  17 | elapsed:    0.6s finished

[2017-11-13 00:47:52] Features: 25/41 -- score: -50.7333998889[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  16 out of  16 | elapsed:    0.6s finished

[2017-11-13 00:47:52] Features: 26/41 -- score: -50.7925769304[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  15 out of  15 | elapsed:    0.6s finished

[2017-11-13 00:47:53] Features: 27/41 -- score: -50.8182544361[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:    0.5s finished

[2017-11-13 00:47:53] Features: 28/41 -- score: -50.8837492159[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  13 out of  13 | elapsed:    0.6s finished

[2017-11-13 00:47:54] Features: 29/41 -- score: -50.9487811048[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.5s finished

[2017-11-13 00:47:54] Features: 30/41 -- score: -51.0153010109[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  11 out of  11 | elapsed:    0.5s finished

[2017-11-13 00:47:55] Features: 31/41 -- score: -51.0914580012[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:    0.4s finished

[2017-11-13 00:47:55] Features: 32/41 -- score: -51.1678287847[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:    0.4s finished

[2017-11-13 00:47:56] Features: 33/41 -- score: -51.2540884582[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:    0.4s finished

[2017-11-13 00:47:56] Features: 34/41 -- score: -51.3523402132[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:    0.3s finished

[2017-11-13 00:47:57] Features: 35/41 -- score: -51.4545664495[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:    0.3s finished

[2017-11-13 00:47:57] Features: 36/41 -- score: -51.5610306069[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.3s finished

[2017-11-13 00:47:57] Features: 37/41 -- score: -51.6776422551[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    0.2s finished

[2017-11-13 00:47:57] Features: 38/41 -- score: -51.8162235084[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.2s finished

[2017-11-13 00:47:57] Features: 39/41 -- score: -51.9941470661[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s finished

[2017-11-13 00:47:58] Features: 40/41 -- score: -52.1689079911[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s finished

[2017-11-13 00:47:58] Features: 41/41 -- score: -52.3437553235



In [26]:

    
df_features_forward_greedy = pd.DataFrame.from_dict(sfs_greedy.get_metric_dict()).T
df_features_forward_greedy









    Out[26]:







  
    
      
      avg_score
      ci_bound
      cv_scores
      feature_idx
      std_dev
      std_err
    
  
  
    
      1
      -58.729
      13.6689
      [-37.4677973768, -33.6521225601, -42.446939286...
      (8,)
      18.404
      6.13466
    
    
      2
      -53.3893
      11.8031
      [-37.4085971788, -33.3538213249, -38.17086898,...
      (8, 14)
      15.8919
      5.29729
    
    
      3
      -52.5487
      11.5306
      [-40.5448889297, -33.7908060261, -36.130671567...
      (8, 4, 14)
      15.525
      5.17499
    
    
      4
      -51.6462
      11.6248
      [-40.9979286367, -34.3073129744, -33.714687223...
      (8, 18, 4, 14)
      15.6518
      5.21727
    
    
      5
      -51.2903
      11.7016
      [-39.0898452124, -33.8116966977, -34.431368342...
      (4, 8, 14, 18, 19)
      15.7553
      5.25175
    
    
      6
      -51.1052
      11.4419
      [-38.0110040218, -33.7795004351, -34.87955241,...
      (4, 8, 14, 18, 19, 29)
      15.4055
      5.13516
    
    
      7
      -50.8921
      11.1997
      [-38.5144679544, -34.5026858283, -35.596701856...
      (4, 8, 14, 16, 18, 19, 29)
      15.0794
      5.02647
    
    
      8
      -50.8261
      11.2098
      [-38.3765655068, -34.3163244505, -35.584880161...
      (4, 37, 8, 14, 16, 18, 19, 29)
      15.093
      5.031
    
    
      9
      -50.7506
      11.2172
      [-38.3022111224, -34.6306287177, -35.889225353...
      (4, 8, 14, 16, 18, 19, 21, 29, 37)
      15.103
      5.03434
    
    
      10
      -50.7005
      11.1924
      [-38.2032136545, -34.3155400654, -36.337505741...
      (4, 8, 14, 16, 18, 19, 21, 29, 31, 37)
      15.0696
      5.0232
    
    
      11
      -50.6449
      11.1133
      [-38.4026056558, -34.3505375776, -36.262873512...
      (4, 8, 9, 14, 16, 18, 19, 21, 29, 31, 37)
      14.9631
      4.9877
    
    
      12
      -50.6057
      11.1371
      [-37.8370282807, -33.9426594389, -36.131244535...
      (4, 8, 9, 14, 16, 18, 19, 21, 29, 31, 32, 37)
      14.9952
      4.99841
    
    
      13
      -50.5972
      11.1849
      [-37.7281145137, -33.7649188255, -36.126451567...
      (4, 8, 9, 14, 16, 18, 19, 21, 29, 31, 32, 34, 37)
      15.0595
      5.01983
    
    
      14
      -50.5179
      11.1246
      [-37.5690466059, -33.6631464507, -36.491006094...
      (4, 8, 9, 14, 16, 18, 19, 21, 26, 29, 31, 32, ...
      14.9784
      4.99279
    
    
      15
      -50.5095
      10.9609
      [-37.267840265, -33.854925969, -36.7490412027,...
      (4, 8, 9, 14, 16, 18, 19, 21, 26, 29, 31, 32, ...
      14.7579
      4.91931
    
    
      16
      -50.5144
      10.9684
      [-37.2149055747, -33.8414719438, -36.735820806...
      (4, 8, 9, 14, 15, 16, 18, 19, 21, 26, 29, 31, ...
      14.7681
      4.92269
    
    
      17
      -50.5349
      10.9509
      [-37.417806343, -33.7374990871, -37.1864269802...
      (4, 8, 9, 14, 15, 16, 18, 19, 21, 26, 29, 31, ...
      14.7445
      4.91483
    
    
      18
      -50.5234
      10.9793
      [-37.3374969057, -33.5459434165, -36.995863156...
      (4, 8, 9, 14, 15, 16, 18, 19, 21, 25, 26, 29, ...
      14.7827
      4.92756
    
    
      19
      -50.5403
      10.8435
      [-37.7709128441, -34.1613830866, -36.907601091...
      (4, 8, 9, 14, 15, 16, 18, 19, 21, 25, 26, 27, ...
      14.5998
      4.86661
    
    
      20
      -50.5578
      10.9664
      [-37.5003482012, -33.9096213377, -36.818504030...
      (4, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25, 26, ...
      14.7653
      4.92176
    
    
      21
      -50.5815
      10.8605
      [-36.9235198403, -34.0273312201, -37.625774458...
      (4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25, 2...
      14.6228
      4.87425
    
    
      22
      -50.5887
      10.6434
      [-38.2239851046, -33.8776648162, -37.180484151...
      (2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25...
      14.3305
      4.77683
    
    
      23
      -50.6244
      10.6484
      [-38.2050458238, -33.8242467268, -37.162471256...
      (2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25...
      14.3372
      4.77906
    
    
      24
      -50.6708
      10.6141
      [-37.9919494418, -33.7048624103, -36.982882272...
      (0, 2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23,...
      14.291
      4.76367
    
    
      25
      -50.7334
      10.6437
      [-37.9927505433, -33.6917842112, -37.167532951...
      (0, 2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23,...
      14.3309
      4.77695
    
    
      26
      -50.7926
      10.655
      [-38.0788122032, -33.9151677521, -37.220912394...
      (0, 2, 4, 7, 8, 9, 14, 15, 16, 17, 18, 19, 21,...
      14.3461
      4.78203
    
    
      27
      -50.8183
      10.5474
      [-38.3421978901, -34.2926283276, -36.904463036...
      (0, 2, 3, 4, 7, 8, 9, 14, 15, 16, 17, 18, 19, ...
      14.2012
      4.73372
    
    
      28
      -50.8837
      10.5988
      [-38.1899632427, -34.4398863025, -36.790505341...
      (0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...
      14.2704
      4.75682
    
    
      29
      -50.9488
      10.6091
      [-38.3673053805, -34.4654642781, -36.812002731...
      (0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...
      14.2842
      4.7614
    
    
      30
      -51.0153
      10.5366
      [-38.283503843, -34.5304392925, -36.9100796902...
      (0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...
      14.1867
      4.7289
    
    
      31
      -51.0915
      10.6862
      [-38.0804753003, -34.3770040789, -37.120038242...
      (0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...
      14.3881
      4.79603
    
    
      32
      -51.1678
      10.7111
      [-38.0837965481, -34.389972527, -37.1455540782...
      (0, 1, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 1...
      14.4215
      4.80718
    
    
      33
      -51.2541
      10.5636
      [-39.7453883135, -33.9928250901, -37.502896145...
      (0, 1, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 1...
      14.223
      4.74101
    
    
      34
      -51.3523
      10.5883
      [-39.7963747212, -34.0273242614, -37.508956594...
      (0, 1, 2, 3, 4, 5, 7, 8, 9, 12, 14, 15, 16, 17...
      14.2563
      4.7521
    
    
      35
      -51.4546
      10.621
      [-39.8843712229, -34.0426710621, -37.473312091...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.3003
      4.76678
    
    
      36
      -51.561
      10.6293
      [-39.8306171226, -33.9311894079, -37.610886776...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.3114
      4.77047
    
    
      37
      -51.6776
      10.6524
      [-39.7812764787, -33.8648868569, -37.706374010...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.3426
      4.78085
    
    
      38
      -51.8162
      10.5379
      [-40.0225616642, -34.6215216995, -38.021994292...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.1884
      4.72948
    
    
      39
      -51.9941
      10.4788
      [-40.7927401909, -34.6496941834, -38.745911429...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 15,...
      14.1088
      4.70293
    
    
      40
      -52.1689
      10.3856
      [-41.7721913235, -34.6381983809, -39.383535235...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14,...
      13.9833
      4.66111
    
    
      41
      -52.3438
      10.8884
      [-41.0492286076, -33.982064226, -39.1396631043...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,...
      14.6603
      4.88676

simple greedy solution:



In [27]:

    
greedy_selected_columns_index=list()

selected_column_indice=df_features_forward_greedy[df_features_forward_greedy['avg_score'] == df_features_forward_greedy['avg_score'].max()]['feature_idx']
greedy_selected_columns_index=selected_column_indice.tolist()
greedy_selected_colname_list = df_before_match_out_training.columns[greedy_selected_columns_index].tolist()
print("Selected columns by simple greedy sequential forward selection are: \n",greedy_selected_colname_list)









    



Selected columns by simple greedy sequential forward selection are: 
 ['buildUpPlayPassing_away', 'IWH', 'chanceCreationCrossing_home', 'B365D', 'WHA', 'buildUpPlayPassing_home', 'defencePressure_home', 'B365H', 'buildUpPlayDribbling_home', 'defenceAggression_away', 'WHH', 'defenceTeamWidth_away', 'buildUpPlaySpeed_home', 'buildUpPlaySpeed_away', 'chanceCreationCrossing_away']



In [28]:

    
lr = linear_model.LinearRegression()
sfs_float = SFS(lr, 
           k_features=df_before_match_out_training.shape[1], 
           forward=True, 
           floating=True, 
           verbose=2,
           scoring='neg_mean_squared_error',
           cv=10)
sfs_float = sfs_float.fit(df_before_match_out_training.as_matrix(), df_before_match_out_trainingtarget.as_matrix())
fig = plot_sfs(sfs_float.get_metric_dict(), kind='std_err')
fig.set_size_inches(18.5, 10.5, forward=True)
plt.title('Sequential Forward Selection (w. StdErr)')
plt.grid()
plt.show()









    



[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  41 out of  41 | elapsed:    0.4s finished

[2017-11-13 00:47:59] Features: 1/41 -- score: -58.7289698381[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  40 out of  40 | elapsed:    0.4s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s finished

[2017-11-13 00:47:59] Features: 2/41 -- score: -53.3892854073[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  39 out of  39 | elapsed:    0.4s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.0s finished

[2017-11-13 00:47:59] Features: 3/41 -- score: -52.5486830747[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  38 out of  38 | elapsed:    0.4s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.0s finished

[2017-11-13 00:48:00] Features: 4/41 -- score: -51.6461824979[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  37 out of  37 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    0.0s finished

[2017-11-13 00:48:00] Features: 5/41 -- score: -51.2902865218[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  36 out of  36 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.1s finished

[2017-11-13 00:48:01] Features: 6/41 -- score: -51.1051971651[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  35 out of  35 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:    0.1s finished

[2017-11-13 00:48:01] Features: 6/41 -- score: -50.8421418182[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  35 out of  35 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:    0.1s finished

[2017-11-13 00:48:02] Features: 7/41 -- score: -50.7796373985[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  34 out of  34 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:    0.1s finished

[2017-11-13 00:48:03] Features: 8/41 -- score: -50.7294001724[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  33 out of  33 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:    0.1s finished

[2017-11-13 00:48:03] Features: 8/41 -- score: -50.5746128643[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  33 out of  33 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:    0.1s finished

[2017-11-13 00:48:04] Features: 9/41 -- score: -50.5311691647[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  32 out of  32 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:    0.1s finished

[2017-11-13 00:48:05] Features: 10/41 -- score: -50.49037586[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  31 out of  31 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:    0.2s finished

[2017-11-13 00:48:05] Features: 11/41 -- score: -50.4782939114[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  30 out of  30 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  11 out of  11 | elapsed:    0.2s finished

[2017-11-13 00:48:06] Features: 12/41 -- score: -50.4661707302[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  29 out of  29 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.2s finished

[2017-11-13 00:48:07] Features: 13/41 -- score: -50.4614059101[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  28 out of  28 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  13 out of  13 | elapsed:    0.3s finished

[2017-11-13 00:48:08] Features: 14/41 -- score: -50.3752532288[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:    0.3s finished

[2017-11-13 00:48:09] Features: 14/41 -- score: -50.3479270261[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:    0.3s finished

[2017-11-13 00:48:10] Features: 14/41 -- score: -50.3353641957[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:    0.3s finished

[2017-11-13 00:48:11] Features: 15/41 -- score: -50.3362224873[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  26 out of  26 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  15 out of  15 | elapsed:    0.3s finished

[2017-11-13 00:48:12] Features: 16/41 -- score: -50.344178819[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  16 out of  16 | elapsed:    0.4s finished

[2017-11-13 00:48:13] Features: 17/41 -- score: -50.3645147044[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  24 out of  24 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  17 out of  17 | elapsed:    0.4s finished

[2017-11-13 00:48:14] Features: 17/41 -- score: -50.3629886102[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  24 out of  24 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  17 out of  17 | elapsed:    0.4s finished

[2017-11-13 00:48:15] Features: 18/41 -- score: -50.3667424162[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  23 out of  23 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:    0.5s finished

[2017-11-13 00:48:16] Features: 19/41 -- score: -50.3786141296[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  22 out of  22 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  19 out of  19 | elapsed:    0.5s finished

[2017-11-13 00:48:17] Features: 20/41 -- score: -50.3939524977[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  21 out of  21 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  20 out of  20 | elapsed:    0.6s finished

[2017-11-13 00:48:18] Features: 21/41 -- score: -50.4312769423[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  20 out of  20 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  21 out of  21 | elapsed:    0.6s finished

[2017-11-13 00:48:19] Features: 22/41 -- score: -50.4887113486[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  19 out of  19 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  22 out of  22 | elapsed:    0.7s finished

[2017-11-13 00:48:21] Features: 23/41 -- score: -50.5455021142[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  23 out of  23 | elapsed:    0.7s finished

[2017-11-13 00:48:22] Features: 23/41 -- score: -50.5407926476[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  18 out of  18 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  23 out of  23 | elapsed:    0.7s finished

[2017-11-13 00:48:23] Features: 24/41 -- score: -50.6017427723[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  17 out of  17 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:    0.9s finished

[2017-11-13 00:48:25] Features: 25/41 -- score: -50.6695389805[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  16 out of  16 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:    0.8s finished

[2017-11-13 00:48:26] Features: 25/41 -- score: -50.6658717205[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  16 out of  16 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  25 out of  25 | elapsed:    0.8s finished

[2017-11-13 00:48:28] Features: 26/41 -- score: -50.7330640587[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  15 out of  15 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  26 out of  26 | elapsed:    0.9s finished

[2017-11-13 00:48:29] Features: 27/41 -- score: -50.8030279956[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:    0.9s finished

[2017-11-13 00:48:30] Features: 27/41 -- score: -50.7654601664[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  14 out of  14 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  27 out of  27 | elapsed:    0.9s finished

[2017-11-13 00:48:32] Features: 28/41 -- score: -50.8287279344[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  13 out of  13 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  28 out of  28 | elapsed:    1.0s finished

[2017-11-13 00:48:33] Features: 29/41 -- score: -50.8939539017[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.5s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  29 out of  29 | elapsed:    1.2s finished

[2017-11-13 00:48:35] Features: 30/41 -- score: -50.9712503962[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  11 out of  11 | elapsed:    0.7s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  30 out of  30 | elapsed:    1.6s finished

[2017-11-13 00:48:37] Features: 31/41 -- score: -51.0484175464[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:    0.6s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  31 out of  31 | elapsed:    1.4s finished

[2017-11-13 00:48:39] Features: 31/41 -- score: -51.0473070694[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:    0.4s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  31 out of  31 | elapsed:    1.3s finished

[2017-11-13 00:48:41] Features: 32/41 -- score: -51.1344295798[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   9 out of   9 | elapsed:    0.4s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  32 out of  32 | elapsed:    1.4s finished

[2017-11-13 00:48:43] Features: 33/41 -- score: -51.2316552168[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   8 out of   8 | elapsed:    0.4s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  33 out of  33 | elapsed:    1.6s finished

[2017-11-13 00:48:45] Features: 34/41 -- score: -51.3411138789[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   7 out of   7 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  34 out of  34 | elapsed:    1.9s finished

[2017-11-13 00:48:47] Features: 35/41 -- score: -51.4608332034[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   6 out of   6 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  35 out of  35 | elapsed:    1.8s finished

[2017-11-13 00:48:49] Features: 36/41 -- score: -51.5775691055[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  36 out of  36 | elapsed:    1.7s finished

[2017-11-13 00:48:51] Features: 36/41 -- score: -51.5610306069[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   5 out of   5 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  36 out of  36 | elapsed:    2.0s finished

[2017-11-13 00:48:54] Features: 37/41 -- score: -51.6776422551[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   4 out of   4 | elapsed:    0.2s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  37 out of  37 | elapsed:    2.6s finished

[2017-11-13 00:48:56] Features: 38/41 -- score: -51.8162235084[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   3 out of   3 | elapsed:    0.3s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  38 out of  38 | elapsed:    2.4s finished

[2017-11-13 00:48:59] Features: 39/41 -- score: -51.9941470661[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   2 out of   2 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  39 out of  39 | elapsed:    2.1s finished

[2017-11-13 00:49:01] Features: 40/41 -- score: -52.1689079911[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  40 out of  40 | elapsed:    2.5s finished

[2017-11-13 00:49:04] Features: 40/41 -- score: -52.1612512491[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s finished
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  40 out of  40 | elapsed:    2.6s finished

[2017-11-13 00:49:06] Features: 41/41 -- score: -52.3437553235



In [29]:

    
df_features_forward_float = pd.DataFrame.from_dict(sfs_float.get_metric_dict()).T
df_features_forward_float









    Out[29]:







  
    
      
      avg_score
      ci_bound
      cv_scores
      feature_idx
      std_dev
      std_err
    
  
  
    
      1
      -58.729
      13.6689
      [-37.4677973768, -33.6521225601, -42.446939286...
      (8,)
      18.404
      6.13466
    
    
      2
      -53.3893
      11.8031
      [-37.4085971788, -33.3538213249, -38.17086898,...
      (8, 14)
      15.8919
      5.29729
    
    
      3
      -52.5487
      11.5306
      [-40.5448889297, -33.7908060261, -36.130671567...
      (8, 4, 14)
      15.525
      5.17499
    
    
      4
      -51.6462
      11.6248
      [-40.9979286367, -34.3073129744, -33.714687223...
      (8, 18, 4, 14)
      15.6518
      5.21727
    
    
      5
      -51.2903
      11.7016
      [-39.0898452124, -33.8116966977, -34.431368342...
      (4, 8, 14, 18, 19)
      15.7553
      5.25175
    
    
      6
      -50.8421
      11.1963
      [-38.5097691075, -34.4637828447, -35.565556063...
      (4, 8, 16, 18, 19, 29)
      15.0748
      5.02494
    
    
      7
      -50.7796
      11.2007
      [-38.3837111765, -34.2938601097, -35.571738831...
      (4, 37, 8, 16, 18, 19, 29)
      15.0808
      5.02693
    
    
      8
      -50.5746
      11.0895
      [-38.5931147694, -34.6772062187, -35.846991528...
      (4, 9, 16, 18, 19, 21, 29, 37)
      14.931
      4.97701
    
    
      9
      -50.5312
      11.0712
      [-38.486072539, -34.4796060666, -36.3341969211...
      (4, 9, 16, 18, 19, 21, 29, 31, 37)
      14.9064
      4.96881
    
    
      10
      -50.4904
      11.0942
      [-37.9098446834, -34.0598054135, -36.211676097...
      (4, 9, 16, 18, 19, 21, 29, 31, 32, 37)
      14.9374
      4.97914
    
    
      11
      -50.4783
      11.1196
      [-37.876265224, -33.9687137999, -36.2417005184...
      (4, 9, 16, 18, 19, 21, 29, 31, 32, 33, 37)
      14.9716
      4.99054
    
    
      12
      -50.4662
      11.1501
      [-37.7902927233, -33.7462770825, -36.074667224...
      (4, 9, 16, 18, 19, 21, 25, 29, 31, 32, 33, 37)
      15.0127
      5.00422
    
    
      13
      -50.4614
      11.185
      [-37.6993005008, -33.6175561993, -36.101592539...
      (4, 9, 16, 18, 19, 21, 25, 29, 31, 32, 33, 34,...
      15.0597
      5.01989
    
    
      14
      -50.3354
      10.6071
      [-37.162695323, -34.4296880226, -37.4500769792...
      (4, 9, 16, 19, 21, 25, 26, 27, 29, 31, 32, 33,...
      14.2815
      4.7605
    
    
      15
      -50.3362
      10.6217
      [-37.0776413792, -34.3946904375, -37.428523195...
      (4, 9, 15, 16, 19, 21, 25, 26, 27, 29, 31, 32,...
      14.3012
      4.76708
    
    
      16
      -50.3442
      10.6757
      [-37.1323843756, -34.2214594917, -37.324255878...
      (4, 9, 15, 16, 19, 21, 25, 26, 27, 29, 31, 32,...
      14.3739
      4.79129
    
    
      17
      -50.363
      10.6773
      [-36.7784028323, -34.4768877999, -37.773264323...
      (4, 7, 9, 15, 16, 18, 19, 21, 25, 26, 27, 29, ...
      14.3761
      4.79205
    
    
      18
      -50.3667
      10.7251
      [-36.8159088189, -34.3254591766, -37.676084400...
      (4, 7, 9, 15, 16, 18, 19, 21, 25, 26, 27, 29, ...
      14.4404
      4.81347
    
    
      19
      -50.3786
      10.5241
      [-37.7495896581, -34.1763476028, -37.216329679...
      (2, 4, 7, 9, 15, 16, 18, 19, 21, 25, 26, 27, 2...
      14.1698
      4.72328
    
    
      20
      -50.394
      10.6502
      [-37.4859018658, -33.9170173776, -37.100808598...
      (2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 25, 26, 2...
      14.3396
      4.77988
    
    
      21
      -50.4313
      10.6488
      [-37.4903287728, -33.8767380398, -37.093130946...
      (2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 25, 26, 2...
      14.3377
      4.77923
    
    
      22
      -50.4887
      10.6815
      [-37.5068026088, -33.8584169043, -37.247595670...
      (2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 25, 26, 2...
      14.3817
      4.7939
    
    
      23
      -50.5408
      10.6247
      [-37.521092032, -33.9753141735, -37.3668478931...
      (2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 24, 25, 2...
      14.3052
      4.7684
    
    
      24
      -50.6017
      10.6146
      [-37.5529634473, -33.9454593459, -37.358597965...
      (2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 24, 25, 2...
      14.2917
      4.7639
    
    
      25
      -50.6659
      10.5087
      [-38.0501228736, -33.5693908449, -37.190543296...
      (0, 2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 24, 25...
      14.1491
      4.71636
    
    
      26
      -50.7331
      10.5333
      [-38.075094857, -33.5614239039, -37.3895592559...
      (0, 2, 4, 7, 9, 15, 16, 18, 19, 21, 23, 24, 25...
      14.1822
      4.72739
    
    
      27
      -50.7655
      10.4942
      [-38.8268682662, -34.0682013394, -37.013885887...
      (0, 2, 3, 4, 7, 9, 15, 16, 17, 18, 19, 21, 23,...
      14.1295
      4.70983
    
    
      28
      -50.8287
      10.4126
      [-38.7448465026, -34.1270113442, -37.14831564,...
      (0, 2, 3, 4, 7, 9, 15, 16, 17, 18, 19, 21, 23,...
      14.0196
      4.67321
    
    
      29
      -50.894
      10.4471
      [-38.615153899, -34.2656497336, -37.0455573175...
      (0, 2, 3, 4, 7, 9, 12, 15, 16, 17, 18, 19, 21,...
      14.0662
      4.68873
    
    
      30
      -50.9713
      10.5985
      [-38.4245009399, -34.1142477321, -37.219373073...
      (0, 2, 3, 4, 7, 9, 12, 15, 16, 17, 18, 19, 20,...
      14.27
      4.75666
    
    
      31
      -51.0473
      10.66
      [-38.1303994507, -34.4249260491, -37.167361629...
      (0, 1, 2, 3, 4, 7, 9, 12, 14, 15, 16, 17, 18, ...
      14.3528
      4.78427
    
    
      32
      -51.1344
      10.5108
      [-39.8009517129, -34.0410975053, -37.520008653...
      (0, 1, 2, 3, 4, 7, 9, 12, 14, 15, 16, 17, 18, ...
      14.152
      4.71732
    
    
      33
      -51.2317
      10.5188
      [-39.730844615, -33.9253466461, -37.6525288368...
      (0, 1, 2, 3, 4, 7, 9, 12, 14, 15, 16, 17, 18, ...
      14.1627
      4.7209
    
    
      34
      -51.3411
      10.5473
      [-39.804217926, -33.9642141989, -37.6346325265...
      (0, 1, 2, 3, 4, 6, 7, 9, 12, 14, 15, 16, 17, 1...
      14.2011
      4.7337
    
    
      35
      -51.4608
      10.5702
      [-39.7562857084, -33.9024652978, -37.731242904...
      (0, 1, 2, 3, 4, 6, 7, 9, 12, 14, 15, 16, 17, 1...
      14.2319
      4.74397
    
    
      36
      -51.561
      10.6293
      [-39.8306171226, -33.9311894079, -37.610886776...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.3114
      4.77047
    
    
      37
      -51.6776
      10.6524
      [-39.7812764787, -33.8648868569, -37.706374010...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.3426
      4.78085
    
    
      38
      -51.8162
      10.5379
      [-40.0225616642, -34.6215216995, -38.021994292...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...
      14.1884
      4.72948
    
    
      39
      -51.9941
      10.4788
      [-40.7927401909, -34.6496941834, -38.745911429...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 15,...
      14.1088
      4.70293
    
    
      40
      -52.1613
      10.9703
      [-40.8142048281, -33.2719365274, -38.866660293...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,...
      14.7705
      4.92351
    
    
      41
      -52.3438
      10.8884
      [-41.0492286076, -33.982064226, -39.1396631043...
      (0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,...
      14.6603
      4.88676

float solution:



In [30]:

    
float_selected_columns_index=list()

selected_column_indice=df_features_forward_float[df_features_forward_float['avg_score'] == df_features_forward_float['avg_score'].max()]['feature_idx']
float_selected_columns_index=selected_column_indice.tolist()
float_selected_colname_list = df_before_match_out_training.columns[float_selected_columns_index].tolist()
print("Selected columns by float greedy sequential forward selection are: \n",float_selected_colname_list)









    



Selected columns by float greedy sequential forward selection are: 
 ['buildUpPlayPassing_away', 'chanceCreationCrossing_home', 'WHA', 'defencePressure_home', 'B365H', 'VCA', 'buildUpPlayDribbling_home', 'LBA', 'defenceAggression_away', 'WHH', 'defenceTeamWidth_away', 'VCD', 'buildUpPlaySpeed_home', 'chanceCreationCrossing_away']

Use the selected columns from both simple greedy and float greedy to use on the outter layer holdout validation

1. Simple Greedy



In [31]:

    
df_before_match_greedy_training=df_before_match_out_training[greedy_selected_colname_list]
df_before_match_greedy_test=df_before_match_out_test[greedy_selected_colname_list]

lr_greedy = linear_model.LinearRegression()
lr_greedy.fit(df_before_match_greedy_training.as_matrix(), df_before_match_out_trainingtarget.as_matrix())
before_match_greedy_pred = lr_greedy.predict(df_before_match_greedy_test.as_matrix())


print(r2_score(df_before_match_out_testtarget.tolist(), before_match_greedy_pred.tolist()))









    



0.419410932893



In [32]:

    
df_before_match_float_training=df_before_match_out_training[float_selected_colname_list]
df_before_match_float_test=df_before_match_out_test[float_selected_colname_list]

lr_greedy = linear_model.LinearRegression()
lr_greedy.fit(df_before_match_float_training.as_matrix(), df_before_match_out_trainingtarget.as_matrix())
before_match_float_pred = lr_greedy.predict(df_before_match_float_test.as_matrix())


print(r2_score(df_before_match_out_testtarget.tolist(), before_match_float_pred.tolist()))









    



0.411118387205

ANN



In [33]:

    
df_before_match.head()









    Out[33]:







  
    
      
      B365A
      BWH
      IWD
      defenceTeamWidth_home
      buildUpPlayPassing_away
      IWA
      LBH
      LBD
      IWH
      chanceCreationCrossing_home
      ...
      defenceTeamWidth_away
      VCD
      buildUpPlaySpeed_home
      defenceAggression_home
      defencePressure_away
      buildUpPlaySpeed_away
      BWA
      chanceCreationCrossing_away
      chanceCreationShooting_home
      possession_home
    
  
  
    
      0
      4.33
      1.80
      3.40
      45
      65
      4.30
      1.80
      3.30
      1.80
      70
      ...
      40
      3.40
      70
      55
      40
      60
      4.00
      70
      70
      57
    
    
      1
      2.10
      3.45
      3.20
      35
      59
      2.20
      3.30
      3.25
      3.20
      70
      ...
      30
      3.25
      70
      70
      30
      70
      2.05
      70
      50
      57
    
    
      2
      4.50
      1.72
      3.50
      30
      70
      4.60
      1.80
      3.30
      1.75
      70
      ...
      70
      3.50
      58
      70
      70
      70
      4.50
      70
      50
      60
    
    
      3
      2.88
      2.45
      3.20
      70
      70
      2.85
      2.38
      3.20
      2.40
      70
      ...
      35
      3.25
      70
      70
      35
      65
      2.75
      70
      70
      51
    
    
      4
      3.75
      2.00
      3.25
      35
      70
      3.40
      2.10
      3.20
      2.10
      70
      ...
      35
      3.25
      60
      70
      35
      55
      3.65
      70
      55
      53
    
  

5 rows × 42 columns



In [34]:

    
print(before_match_attributes)









    



['B365A', 'BWH', 'IWD', 'defenceTeamWidth_home', 'buildUpPlayPassing_away', 'IWA', 'LBH', 'LBD', 'IWH', 'chanceCreationCrossing_home', 'WHD', 'buildUpPlayPositioningClass_home_isOrganised', 'BWD', 'buildUpPlayDribbling_away', 'B365D', 'stage', 'WHA', 'chanceCreationPositioningClass_home_isOrganised', 'buildUpPlayPassing_home', 'defencePressure_home', 'buildUpPlayPositioningClass_away_isOrganised', 'B365H', 'chanceCreationPositioningClass_away_isOrganised', 'chanceCreationPassing_home', 'chanceCreationShooting_away', 'VCA', 'buildUpPlayDribbling_home', 'LBA', 'chanceCreationPassing_away', 'defenceAggression_away', 'VCH', 'WHH', 'defenceTeamWidth_away', 'VCD', 'buildUpPlaySpeed_home', 'defenceAggression_home', 'defencePressure_away', 'buildUpPlaySpeed_away', 'BWA', 'chanceCreationCrossing_away', 'chanceCreationShooting_home']



In [35]:

    
#before_match_attributes = ['LBH']



In [36]:

    
# ANN
from sklearn.neural_network import MLPRegressor
from sklearn.model_selection import train_test_split

df = df_before_match

train_df, test_df = train_test_split(df, test_size=0.2)

trainX = train_df[before_match_attributes]
trainY = train_df['possession_home']

testX = test_df[before_match_attributes]
testY = test_df['possession_home']

fnn = MLPRegressor(solver='adam', alpha=1e-2,
                    hidden_layer_sizes=(64, 32, ),
                    random_state=1,
                    nesterovs_momentum=False,
                    momentum=0.99,
                    early_stopping=False,
                    activation='logistic',
                    batch_size=16,
                    warm_start=False,
                    max_iter=3000)

#fnn = MLPRegressor(solver='adam', alpha=1e-3,
#                    hidden_layer_sizes=(256,128,64,32,),
#                    random_state=1,
#                    nesterovs_momentum=False,
#                    momentum=0.99,
#                    early_stopping=False,
#                    activation='logistic',
#                    batch_size=32,
#                    warm_start=False,
#                  max_iter=3000)

model = fnn.fit(trainX, trainY)



In [37]:

    
print(model)
print(model.loss_curve_)
print(model.n_layers_)









    



MLPRegressor(activation='logistic', alpha=0.01, batch_size=16, beta_1=0.9,
       beta_2=0.999, early_stopping=False, epsilon=1e-08,
       hidden_layer_sizes=(64, 32), learning_rate='constant',
       learning_rate_init=0.001, max_iter=3000, momentum=0.99,
       nesterovs_momentum=False, power_t=0.5, random_state=1, shuffle=True,
       solver='adam', tol=0.0001, validation_fraction=0.1, verbose=False,
       warm_start=False)
[1200.1425832479626, 1030.8465655639943, 901.73341935577207, 790.65377009995586, 692.09592681624565, 604.15318972276862, 525.70986952476676, 454.08097856591775, 382.89424855528421, 324.24603921633218, 274.68048144383471, 228.56603446217903, 187.91930165888178, 156.02846421500601, 130.29681098832424, 109.59223502768197, 93.0250131856953, 79.867372008856222, 69.569877191645929, 61.631809712298313, 55.61156768653553, 51.134339526499694, 47.881628534513965, 45.550181772278329, 43.929284102276391, 42.822140436939144, 42.101093626371437, 41.641600736998875, 41.35119329378, 41.174582803483325, 40.876558131194166, 38.687902362345525, 35.074774448373084, 33.364493519753438, 31.424531514217403, 30.243282151138988, 28.886210078127032, 28.601150702207452, 27.555117363406492, 27.430190503832666, 26.863470417340213, 26.181426457897174, 25.941187239312946, 25.93225766136047, 25.760779857739738, 25.217683265513465, 25.280890686112567, 25.332715373120607, 25.207470006542547, 25.310669850937554, 24.989191694026729, 24.952350630070903, 25.081061309572842, 24.782444233306872, 24.528562478035116, 24.540893107118972, 24.331748563943982, 24.439230906599427, 24.694451405380704, 24.6375830193345]
4



In [38]:

    
#model.predict(testX) - testY
y_pred = model.predict(testX)
y_true = testY
mean_squared_error(y_true, y_pred)









    Out[38]:





49.099689995701375



In [39]:

    
print("Predicted: " + str(model.predict(testX.iloc[470].values.reshape(1, -1))) + ", actual: " + str(testY.iloc[470]))

print("Predicted: " + str(model.predict(testX.iloc[123].values.reshape(1, -1))) + ", actual: " + str(testY.iloc[123]))

print("Predicted: " + str(model.predict(testX.iloc[321].values.reshape(1, -1))) + ", actual: " + str(testY.iloc[321]))

print("Predicted: " + str(model.predict(testX.iloc[420].values.reshape(1, -1))) + ", actual: " + str(testY.iloc[420]))









    



Predicted: [ 46.6441023], actual: 41
Predicted: [ 52.60814435], actual: 36
Predicted: [ 57.0227647], actual: 55
Predicted: [ 43.43182945], actual: 41



In [56]:

    
from sklearn.model_selection import KFold

linear_features = ['buildUpPlayPassing_home', 'defenceTeamWidth_away', 'B365H',
                   'buildUpPlayDribbling_home', 'buildUpPlaySpeed_home', 'chanceCreationShooting_away',
                   'IWA', 'chanceCreationCrossing_home', 'buildUpPlaySpeed_away',
                   'buildUpPlayPassing_away', 'defenceAggression_away', 'defencePressure_home']


X_fnn = df[before_match_attributes]
X_lin = df[linear_features]
y = df['possession_home']

K = 10
v = K - 1


kf = KFold(n_splits=K, shuffle=True)

fnn_tests = []
lin_tests = []
avg_tests = []

for train_index, test_index in kf.split(X_fnn):
   
    X_train_fnn, X_test_fnn = X_fnn.iloc[train_index], X_fnn.iloc[test_index]
    X_train_lin, X_test_lin = X_lin.iloc[train_index], X_lin.iloc[test_index]
    y_train, y_test = y.iloc[train_index], y.iloc[test_index]
    fnn_fit = fnn.fit(X_train_fnn, y_train)
    y_pred_fnn = fnn_fit.predict(X_test_fnn)
    mse_fnn = mean_squared_error(y_test, y_pred_fnn)
    print("FNN: " + str(mse_fnn))
    fnn_tests.append(mse_fnn)
    
    lin_model = linear_model.LinearRegression()
    lin_fit = lin_model.fit(X_train_lin, y_train)
    y_pred_lin = lin_fit.predict(X_test_lin)
    mse_lin = mean_squared_error(y_test, y_pred_lin)
    print("LIN: " + str(mse_lin))
    lin_tests.append(mse_lin)
    
    y_train_avg = sum(y_train) / len(y_train)
    y_pred_const = np.array([y_train_avg] * len(y_test))
    mse_const = mean_squared_error(y_test, y_pred_const)
    print("AVG: " + str(mse_const))
    avg_tests.append(mse_const)
    print("AVG prediction: " + str(y_train_avg))

print("FNN CV MSE: " + str(np.mean(fnn_tests)))
print("Linear model CV MSE: " + str(np.mean(lin_tests)))
print("Const AVG CV MSE: " + str(np.mean(avg_tests)))









    



FNN: 53.7428261879
LIN: 51.602221394
AVG: 83.3800597243
AVG prediction: 51.6256386438
FNN: 58.1194347358
LIN: 58.7772897795
AVG: 89.1005939237
AVG prediction: 51.709707385
FNN: 47.9960029345
LIN: 47.1463835481
AVG: 78.1850063797
AVG prediction: 51.5777984208
FNN: 46.0596158855
LIN: 39.2022045417
AVG: 76.8370232991
AVG prediction: 51.6750232126
FNN: 49.4881387847
LIN: 49.4301577786
AVG: 85.0996755172
AVG prediction: 51.6490250696
FNN: 56.1728310064
LIN: 56.9155555269
AVG: 85.1482352669
AVG prediction: 51.6592386258
FNN: 49.0539094851
LIN: 51.8051889093
AVG: 94.7636585288
AVG prediction: 51.6360259981
FNN: 46.8305870129
LIN: 45.4211481654
AVG: 80.5145389057
AVG prediction: 51.6573816156
FNN: 42.6963958484
LIN: 43.4951065703
AVG: 71.9669609269
AVG prediction: 51.5766016713
FNN: 55.1506126796
LIN: 49.9174763131
AVG: 72.2146229113
AVG prediction: 51.6063138347
FNN CV MSE: 50.5310354561
Linear model CV MSE: 49.3712732527
Const AVG CV MSE: 81.7210375384



In [58]:

    
# Performance comparison
# K=10, v=9
alpha = 0.05
zs = np.array(fnn_tests) - np.array(lin_tests)
z_bar = np.mean(zs)
std_z_bar = np.std(zs)

from scipy.stats import t

theta_l = z_bar + std_z_bar * t.ppf(alpha/2, v)
theta_u = z_bar + std_z_bar * t.ppf(1 - alpha/2, v)
print("z: " + str(z_bar))
print("theta_l: " + str(theta_l))
print("theta_u: " + str(theta_u))









    



z: 1.15976220341
theta_l: -5.13558727048
theta_u: 7.45511167731

	avg_score	ci_bound	cv_scores	feature_idx	std_dev	std_err
1	-58.729	13.6689	[-37.4677973768, -33.6521225601, -42.446939286...	(8,)	18.404	6.13466
2	-53.3893	11.8031	[-37.4085971788, -33.3538213249, -38.17086898,...	(8, 14)	15.8919	5.29729
3	-52.5487	11.5306	[-40.5448889297, -33.7908060261, -36.130671567...	(8, 4, 14)	15.525	5.17499
4	-51.6462	11.6248	[-40.9979286367, -34.3073129744, -33.714687223...	(8, 18, 4, 14)	15.6518	5.21727
5	-51.2903	11.7016	[-39.0898452124, -33.8116966977, -34.431368342...	(4, 8, 14, 18, 19)	15.7553	5.25175
6	-51.1052	11.4419	[-38.0110040218, -33.7795004351, -34.87955241,...	(4, 8, 14, 18, 19, 29)	15.4055	5.13516
7	-50.8921	11.1997	[-38.5144679544, -34.5026858283, -35.596701856...	(4, 8, 14, 16, 18, 19, 29)	15.0794	5.02647
8	-50.8261	11.2098	[-38.3765655068, -34.3163244505, -35.584880161...	(4, 37, 8, 14, 16, 18, 19, 29)	15.093	5.031
9	-50.7506	11.2172	[-38.3022111224, -34.6306287177, -35.889225353...	(4, 8, 14, 16, 18, 19, 21, 29, 37)	15.103	5.03434
10	-50.7005	11.1924	[-38.2032136545, -34.3155400654, -36.337505741...	(4, 8, 14, 16, 18, 19, 21, 29, 31, 37)	15.0696	5.0232
11	-50.6449	11.1133	[-38.4026056558, -34.3505375776, -36.262873512...	(4, 8, 9, 14, 16, 18, 19, 21, 29, 31, 37)	14.9631	4.9877
12	-50.6057	11.1371	[-37.8370282807, -33.9426594389, -36.131244535...	(4, 8, 9, 14, 16, 18, 19, 21, 29, 31, 32, 37)	14.9952	4.99841
13	-50.5972	11.1849	[-37.7281145137, -33.7649188255, -36.126451567...	(4, 8, 9, 14, 16, 18, 19, 21, 29, 31, 32, 34, 37)	15.0595	5.01983
14	-50.5179	11.1246	[-37.5690466059, -33.6631464507, -36.491006094...	(4, 8, 9, 14, 16, 18, 19, 21, 26, 29, 31, 32, ...	14.9784	4.99279
15	-50.5095	10.9609	[-37.267840265, -33.854925969, -36.7490412027,...	(4, 8, 9, 14, 16, 18, 19, 21, 26, 29, 31, 32, ...	14.7579	4.91931
16	-50.5144	10.9684	[-37.2149055747, -33.8414719438, -36.735820806...	(4, 8, 9, 14, 15, 16, 18, 19, 21, 26, 29, 31, ...	14.7681	4.92269
17	-50.5349	10.9509	[-37.417806343, -33.7374990871, -37.1864269802...	(4, 8, 9, 14, 15, 16, 18, 19, 21, 26, 29, 31, ...	14.7445	4.91483
18	-50.5234	10.9793	[-37.3374969057, -33.5459434165, -36.995863156...	(4, 8, 9, 14, 15, 16, 18, 19, 21, 25, 26, 29, ...	14.7827	4.92756
19	-50.5403	10.8435	[-37.7709128441, -34.1613830866, -36.907601091...	(4, 8, 9, 14, 15, 16, 18, 19, 21, 25, 26, 27, ...	14.5998	4.86661
20	-50.5578	10.9664	[-37.5003482012, -33.9096213377, -36.818504030...	(4, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25, 26, ...	14.7653	4.92176
21	-50.5815	10.8605	[-36.9235198403, -34.0273312201, -37.625774458...	(4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25, 2...	14.6228	4.87425
22	-50.5887	10.6434	[-38.2239851046, -33.8776648162, -37.180484151...	(2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25...	14.3305	4.77683
23	-50.6244	10.6484	[-38.2050458238, -33.8242467268, -37.162471256...	(2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23, 25...	14.3372	4.77906
24	-50.6708	10.6141	[-37.9919494418, -33.7048624103, -36.982882272...	(0, 2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23,...	14.291	4.76367
25	-50.7334	10.6437	[-37.9927505433, -33.6917842112, -37.167532951...	(0, 2, 4, 7, 8, 9, 14, 15, 16, 18, 19, 21, 23,...	14.3309	4.77695
26	-50.7926	10.655	[-38.0788122032, -33.9151677521, -37.220912394...	(0, 2, 4, 7, 8, 9, 14, 15, 16, 17, 18, 19, 21,...	14.3461	4.78203
27	-50.8183	10.5474	[-38.3421978901, -34.2926283276, -36.904463036...	(0, 2, 3, 4, 7, 8, 9, 14, 15, 16, 17, 18, 19, ...	14.2012	4.73372
28	-50.8837	10.5988	[-38.1899632427, -34.4398863025, -36.790505341...	(0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...	14.2704	4.75682
29	-50.9488	10.6091	[-38.3673053805, -34.4654642781, -36.812002731...	(0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...	14.2842	4.7614
30	-51.0153	10.5366	[-38.283503843, -34.5304392925, -36.9100796902...	(0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...	14.1867	4.7289
31	-51.0915	10.6862	[-38.0804753003, -34.3770040789, -37.120038242...	(0, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 18, ...	14.3881	4.79603
32	-51.1678	10.7111	[-38.0837965481, -34.389972527, -37.1455540782...	(0, 1, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 1...	14.4215	4.80718
33	-51.2541	10.5636	[-39.7453883135, -33.9928250901, -37.502896145...	(0, 1, 2, 3, 4, 7, 8, 9, 12, 14, 15, 16, 17, 1...	14.223	4.74101
34	-51.3523	10.5883	[-39.7963747212, -34.0273242614, -37.508956594...	(0, 1, 2, 3, 4, 5, 7, 8, 9, 12, 14, 15, 16, 17...	14.2563	4.7521
35	-51.4546	10.621	[-39.8843712229, -34.0426710621, -37.473312091...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...	14.3003	4.76678
36	-51.561	10.6293	[-39.8306171226, -33.9311894079, -37.610886776...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...	14.3114	4.77047
37	-51.6776	10.6524	[-39.7812764787, -33.8648868569, -37.706374010...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...	14.3426	4.78085
38	-51.8162	10.5379	[-40.0225616642, -34.6215216995, -38.021994292...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 14, 15, 16,...	14.1884	4.72948
39	-51.9941	10.4788	[-40.7927401909, -34.6496941834, -38.745911429...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 15,...	14.1088	4.70293
40	-52.1689	10.3856	[-41.7721913235, -34.6381983809, -39.383535235...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14,...	13.9833	4.66111
41	-52.3438	10.8884	[-41.0492286076, -33.982064226, -39.1396631043...	(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13,...	14.6603	4.88676

	B365A	BWH	IWD	defenceTeamWidth_home	buildUpPlayPassing_away	IWA	LBH	LBD	IWH	chanceCreationCrossing_home	...	defenceTeamWidth_away	VCD	buildUpPlaySpeed_home	defenceAggression_home	defencePressure_away	buildUpPlaySpeed_away	BWA	chanceCreationCrossing_away	chanceCreationShooting_home	possession_home
0	4.33	1.80	3.40	45	65	4.30	1.80	3.30	1.80	70	...	40	3.40	70	55	40	60	4.00	70	70	57
1	2.10	3.45	3.20	35	59	2.20	3.30	3.25	3.20	70	...	30	3.25	70	70	30	70	2.05	70	50	57
2	4.50	1.72	3.50	30	70	4.60	1.80	3.30	1.75	70	...	70	3.50	58	70	70	70	4.50	70	50	60
3	2.88	2.45	3.20	70	70	2.85	2.38	3.20	2.40	70	...	35	3.25	70	70	35	65	2.75	70	70	51
4	3.75	2.00	3.25	35	70	3.40	2.10	3.20	2.10	70	...	35	3.25	60	70	35	55	3.65	70	55	53