TF-Core - eLU - Spitzer Calibration Data

This script show a simple example of using [tf.contrib.learn][1] library to create our model.

The code is divided in following steps:

Load CSVs data
Continuous features
Converting Data into Tensors
Selecting and Engineering Features for the Model
Defining The Regression Model
Training and Evaluating Our Model
Predicting output for test data



In [17]:

    
import pandas as pd
import numpy as np
import tensorflow as tf
tf.logging.set_verbosity(tf.logging.ERROR)

import warnings
warnings.filterwarnings("ignore")

%matplotlib inline
from matplotlib import pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, MinMaxScaler, minmax_scale

from sklearn.metrics import r2_score

from tqdm import tqdm_notebook

from time import time
start0 = time()
plt.rcParams['figure.dpi'] = 300



In [18]:

    
# Common imports
import numpy as np
import os

# to make this notebook's output stable across runs
def reset_graph(seed=42):
    tf.reset_default_graph()
    tf.set_random_seed(seed)
    np.random.seed(seed)

# To plot pretty figures
# %matplotlib inline
# import matplotlib
# import matplotlib.pyplot as plt
# plt.rcParams['axes.labelsize'] = 14
# plt.rcParams['xtick.labelsize'] = 12
# plt.rcParams['ytick.labelsize'] = 12

# Where to save the figures
# PROJECT_ROOT_DIR = "."
# CHAPTER_ID = "ann"

# def save_fig(fig_id, tight_layout=True):
#     path = os.path.join(PROJECT_ROOT_DIR, "images", CHAPTER_ID, fig_id + ".png")
#     print("Saving figure", fig_id)
#     if tight_layout:
#         plt.tight_layout()
#     plt.savefig(path, format='png', dpi=300)

Load CSVs data

df_train_ori = pd.read_csv('train.csv') df_test_ori = pd.read_csv('test.csv')



In [19]:

    
nSkip = 20
spitzerDataRaw  = pd.read_csv('pmap_ch2_0p1s_x4_rmulti_s3_7.csv')



In [20]:

    
PLDpixels = pd.DataFrame({key:spitzerDataRaw[key] for key in spitzerDataRaw.columns.values if 'pix' in key})
PLDpixels









    Out[20]:






  
    
      
      pix1
      pix2
      pix3
      pix4
      pix5
      pix6
      pix7
      pix8
      pix9
    
  
  
    
      0
      577.447021
      3465.876709
      1118.598145
      550.165466
      2460.376953
      994.374207
      141.741592
      521.385254
      694.330688
    
    
      1
      569.863098
      3387.739258
      1087.530762
      556.717407
      2552.070557
      1021.892700
      134.061081
      511.347778
      666.346069
    
    
      2
      552.641235
      3405.411377
      1082.131104
      558.981445
      2560.040771
      1058.485352
      146.488220
      513.809570
      691.019653
    
    
      3
      571.821167
      3340.533691
      1073.962036
      568.324768
      2643.155273
      1024.431641
      147.687546
      525.451538
      727.683472
    
    
      4
      538.292114
      3248.569336
      1021.301208
      548.598145
      2691.563965
      1066.199707
      154.170990
      541.407532
      718.537537
    
    
      5
      553.332214
      3183.050293
      1026.863281
      578.003784
      2679.043457
      1085.607422
      154.789886
      517.848328
      722.456909
    
    
      6
      541.202332
      3137.938232
      1035.948364
      589.566528
      2743.923584
      1072.393555
      158.401291
      539.485718
      724.184937
    
    
      7
      547.699829
      3057.428467
      1034.417603
      590.498108
      2818.238770
      1118.098633
      167.132004
      540.160950
      760.767334
    
    
      8
      534.724976
      3015.335205
      1007.406494
      574.792358
      2876.417236
      1141.010498
      158.403992
      527.611633
      765.304077
    
    
      9
      529.423950
      3010.478027
      973.444946
      578.736755
      2875.608887
      1106.352905
      163.936325
      540.751587
      746.721191
    
    
      10
      543.917847
      2959.855225
      1002.942017
      600.163574
      2921.819580
      1124.441772
      153.599976
      548.917603
      735.936890
    
    
      11
      415.543091
      1085.521729
      446.313171
      1641.077881
      4778.175781
      319.484497
      418.018860
      810.050232
      567.318176
    
    
      12
      386.626038
      990.775635
      455.084015
      1653.742676
      4831.918457
      318.239410
      428.828278
      836.078308
      576.660156
    
    
      13
      352.204071
      909.555847
      448.546478
      1697.422363
      4942.354004
      315.904175
      434.170258
      822.891968
      576.275024
    
    
      14
      356.336823
      830.570557
      446.085022
      1725.667725
      4915.351074
      312.517944
      455.358154
      826.022034
      565.244507
    
    
      15
      346.803467
      827.159546
      445.986084
      1783.878418
      4926.551270
      299.384399
      454.638336
      856.649109
      572.549500
    
    
      16
      319.600525
      778.882324
      452.693604
      1766.618774
      4900.985352
      295.125763
      455.190002
      873.377808
      581.590820
    
    
      17
      318.834503
      772.429138
      450.526428
      1804.441284
      4888.824219
      292.672455
      460.712219
      873.199463
      557.806458
    
    
      18
      304.534882
      747.088501
      433.972229
      1812.826172
      4949.355469
      292.384491
      459.964996
      883.855896
      556.133789
    
    
      19
      297.334167
      711.377441
      427.756256
      1824.887939
      4977.011230
      288.222961
      442.602753
      883.110352
      578.062012
    
    
      20
      291.785248
      696.303345
      434.420410
      1808.774414
      4967.067383
      288.906006
      465.871857
      848.896790
      584.660889
    
    
      21
      275.729004
      684.407776
      436.270630
      1839.743286
      4958.617188
      292.503906
      473.751465
      874.837891
      576.924805
    
    
      22
      267.109802
      668.616150
      428.174561
      1800.920166
      5008.872559
      289.184784
      474.868591
      872.165161
      579.913147
    
    
      23
      272.932220
      664.508301
      437.244934
      1776.687744
      5002.353027
      288.002045
      470.427643
      882.427673
      600.059509
    
    
      24
      269.932495
      641.567139
      432.846771
      1789.313354
      5064.815918
      303.351807
      484.210266
      894.807068
      602.440247
    
    
      25
      265.979431
      648.039856
      432.684937
      1755.147705
      5088.369141
      288.072937
      482.552338
      882.221191
      592.758179
    
    
      26
      564.362610
      1541.839600
      431.914764
      1446.160400
      4362.256348
      345.700531
      401.668152
      821.724915
      520.839478
    
    
      27
      506.248077
      1415.009277
      443.126648
      1487.735840
      4414.598633
      347.017731
      393.479889
      795.969360
      549.537842
    
    
      28
      515.429810
      1315.711914
      448.500977
      1547.112671
      4509.289062
      323.252441
      394.831543
      821.549072
      529.604004
    
    
      29
      476.979919
      1244.519043
      429.070221
      1599.996094
      4529.244629
      330.517914
      398.165985
      820.598389
      528.216064
    
    
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    
    
      785255
      475.147705
      2842.766357
      591.782959
      633.689026
      3837.167969
      771.729736
      238.382645
      637.508911
      712.187988
    
    
      785256
      475.464050
      2835.344727
      583.647827
      640.641846
      3873.379639
      750.115234
      248.780396
      651.230103
      716.854736
    
    
      785257
      481.847534
      2817.340820
      578.740479
      637.754578
      3877.218262
      781.961548
      241.087128
      650.161316
      733.283325
    
    
      785258
      465.787201
      2804.904053
      569.343262
      647.821289
      3857.104492
      767.500122
      240.323898
      628.991760
      726.848083
    
    
      785259
      473.557251
      2804.657471
      561.592529
      633.638123
      3901.351562
      773.559692
      231.294861
      641.328003
      733.939087
    
    
      785260
      476.991699
      2791.731445
      579.058228
      643.239990
      3923.991455
      765.078247
      235.408035
      661.097107
      718.653198
    
    
      785261
      476.635651
      2750.070068
      563.718689
      635.713318
      3949.495361
      772.300659
      246.790222
      632.139221
      730.945618
    
    
      785262
      481.483948
      2736.324219
      567.592407
      660.586548
      3944.277100
      766.693787
      241.954529
      633.959961
      734.623413
    
    
      785263
      472.956024
      2736.230225
      562.863098
      659.623413
      4004.276367
      764.286011
      239.283524
      660.717346
      713.998047
    
    
      785264
      466.867065
      2732.458008
      569.005798
      666.330994
      4024.462891
      769.562134
      245.346207
      663.933350
      725.252869
    
    
      785265
      480.046295
      2715.175537
      557.256165
      645.355225
      4037.946533
      742.447998
      258.840576
      654.178589
      706.549988
    
    
      785266
      475.449463
      2769.531494
      536.281189
      655.740417
      4031.584229
      730.941650
      255.209244
      658.127136
      719.966553
    
    
      785267
      502.837372
      2716.243408
      554.105591
      677.264648
      3980.764404
      730.059265
      259.389648
      667.066284
      708.509277
    
    
      785268
      480.399200
      2769.769531
      562.770813
      679.593140
      3983.617920
      712.878052
      243.168762
      654.615112
      689.989075
    
    
      785269
      480.807373
      2759.118408
      561.283997
      699.134888
      4009.483398
      721.416626
      262.486328
      654.512329
      713.022461
    
    
      785270
      523.245300
      2958.067139
      543.905029
      697.710938
      3776.409424
      666.272888
      263.427338
      651.643799
      670.494263
    
    
      785271
      495.875732
      2955.598145
      552.345215
      657.297363
      3747.275879
      703.948242
      250.976898
      683.805420
      657.604126
    
    
      785272
      516.237366
      2964.318604
      577.262634
      668.211304
      3793.730225
      709.129517
      256.492645
      667.471802
      683.404236
    
    
      785273
      501.681335
      2919.767578
      550.606995
      638.708374
      3802.873535
      717.027649
      250.493073
      644.383301
      702.473267
    
    
      785274
      510.000214
      2995.055420
      551.373047
      660.349304
      3745.079590
      722.202332
      247.733948
      642.584961
      695.485779
    
    
      785275
      521.931641
      3013.854004
      554.299988
      631.953613
      3710.329346
      728.134094
      243.888962
      643.002563
      691.181763
    
    
      785276
      495.807709
      2989.654053
      585.109985
      639.402100
      3714.461670
      749.341003
      235.572479
      646.545593
      676.286865
    
    
      785277
      512.751160
      2958.394531
      582.242310
      652.946533
      3707.906250
      729.102295
      234.076538
      652.182556
      688.491699
    
    
      785278
      504.160034
      2989.499268
      587.763611
      658.194824
      3695.002930
      736.696594
      231.465927
      651.476196
      681.635620
    
    
      785279
      516.517090
      2976.608643
      586.910278
      637.642395
      3714.598877
      752.355835
      235.402084
      637.375977
      700.220703
    
    
      785280
      506.839264
      3055.761230
      596.532349
      647.959351
      3610.522217
      731.254211
      234.202530
      658.012695
      675.393433
    
    
      785281
      495.875641
      3055.138428
      590.330444
      623.768982
      3686.792236
      732.337524
      234.055603
      654.529114
      690.857361
    
    
      785282
      510.378479
      3120.925781
      585.978516
      623.555176
      3621.902100
      728.794128
      230.795486
      644.155518
      694.674377
    
    
      785283
      520.521362
      3118.752197
      582.462891
      632.206604
      3600.458984
      756.452515
      224.326355
      641.459534
      696.739502
    
    
      785284
      520.870972
      3144.105225
      609.523743
      617.985229
      3561.716309
      752.712402
      225.764816
      620.627991
      680.224121
    
  

785285 rows × 9 columns



In [21]:

    
PLDnorm = np.sum(np.array(PLDpixels),axis=1)



In [22]:

    
PLDpixels = (PLDpixels.T / PLDnorm).T
PLDpixels









    Out[22]:






  
    
      
      pix1
      pix2
      pix3
      pix4
      pix5
      pix6
      pix7
      pix8
      pix9
    
  
  
    
      0
      0.054868
      0.329321
      0.106287
      0.052276
      0.233781
      0.094484
      0.013468
      0.049541
      0.065974
    
    
      1
      0.054337
      0.323024
      0.103697
      0.053084
      0.243342
      0.097438
      0.012783
      0.048758
      0.063537
    
    
      2
      0.052289
      0.322207
      0.102387
      0.052889
      0.242221
      0.100150
      0.013860
      0.048615
      0.065382
    
    
      3
      0.053828
      0.314461
      0.101097
      0.053499
      0.248813
      0.096435
      0.013903
      0.049463
      0.068500
    
    
      4
      0.051126
      0.308546
      0.097002
      0.052105
      0.255642
      0.101267
      0.014643
      0.051422
      0.068246
    
    
      5
      0.052693
      0.303119
      0.097787
      0.055043
      0.255123
      0.103381
      0.014740
      0.049314
      0.068799
    
    
      6
      0.051333
      0.297631
      0.098259
      0.055920
      0.260259
      0.101716
      0.015024
      0.051170
      0.068688
    
    
      7
      0.051502
      0.287502
      0.097271
      0.055527
      0.265011
      0.105139
      0.015716
      0.050794
      0.071538
    
    
      8
      0.050441
      0.284439
      0.095029
      0.054221
      0.271334
      0.107632
      0.014942
      0.049770
      0.072192
    
    
      9
      0.050299
      0.286019
      0.092485
      0.054984
      0.273205
      0.105112
      0.015575
      0.051376
      0.070944
    
    
      10
      0.051354
      0.279453
      0.094692
      0.056664
      0.275862
      0.106164
      0.014502
      0.051826
      0.069483
    
    
      11
      0.039645
      0.103565
      0.042581
      0.156569
      0.455867
      0.030481
      0.039882
      0.077284
      0.054126
    
    
      12
      0.036899
      0.094558
      0.043433
      0.157831
      0.461151
      0.030372
      0.040927
      0.079794
      0.055036
    
    
      13
      0.033545
      0.086630
      0.042721
      0.161670
      0.470731
      0.030088
      0.041352
      0.078376
      0.054887
    
    
      14
      0.034154
      0.079609
      0.042756
      0.165402
      0.471128
      0.029954
      0.043645
      0.079173
      0.054178
    
    
      15
      0.032986
      0.078675
      0.042420
      0.169673
      0.468588
      0.028476
      0.043243
      0.081480
      0.054458
    
    
      16
      0.030660
      0.074720
      0.043428
      0.169475
      0.470161
      0.028312
      0.043667
      0.083785
      0.055793
    
    
      17
      0.030600
      0.074133
      0.043239
      0.173180
      0.469202
      0.028089
      0.044217
      0.083805
      0.053535
    
    
      18
      0.029170
      0.071559
      0.041568
      0.173640
      0.474071
      0.028006
      0.044057
      0.084660
      0.053269
    
    
      19
      0.028507
      0.068203
      0.041011
      0.174959
      0.477166
      0.027633
      0.042434
      0.084667
      0.055421
    
    
      20
      0.028092
      0.067038
      0.041825
      0.174144
      0.478215
      0.027815
      0.044853
      0.081729
      0.056289
    
    
      21
      0.026480
      0.065728
      0.041898
      0.176681
      0.476205
      0.028091
      0.045497
      0.084016
      0.055405
    
    
      22
      0.025709
      0.064353
      0.041211
      0.173335
      0.482094
      0.027833
      0.045705
      0.083944
      0.055815
    
    
      23
      0.026257
      0.063928
      0.042064
      0.170923
      0.481243
      0.027707
      0.045257
      0.084893
      0.057728
    
    
      24
      0.025749
      0.061199
      0.041289
      0.170683
      0.483133
      0.028937
      0.046189
      0.085356
      0.057467
    
    
      25
      0.025487
      0.062098
      0.041461
      0.168185
      0.487587
      0.027604
      0.046240
      0.084538
      0.056800
    
    
      26
      0.054076
      0.147736
      0.041385
      0.138568
      0.417982
      0.033124
      0.038487
      0.078736
      0.049906
    
    
      27
      0.048900
      0.136680
      0.042803
      0.143705
      0.426419
      0.033519
      0.038007
      0.076885
      0.053081
    
    
      28
      0.049535
      0.126447
      0.043103
      0.148685
      0.433365
      0.031066
      0.037945
      0.078955
      0.050898
    
    
      29
      0.046052
      0.120159
      0.041427
      0.154480
      0.437299
      0.031912
      0.038443
      0.079229
      0.050999
    
    
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
      ...
    
    
      785255
      0.044239
      0.264681
      0.055099
      0.059001
      0.357266
      0.071853
      0.022195
      0.059356
      0.066309
    
    
      785256
      0.044125
      0.263130
      0.054165
      0.059454
      0.359463
      0.069613
      0.023088
      0.060436
      0.066527
    
    
      785257
      0.044618
      0.260880
      0.053590
      0.059055
      0.359022
      0.072408
      0.022324
      0.060203
      0.067900
    
    
      785258
      0.043496
      0.261929
      0.053167
      0.060495
      0.360187
      0.071671
      0.022442
      0.058737
      0.067875
    
    
      785259
      0.044032
      0.260779
      0.052217
      0.058916
      0.362750
      0.071926
      0.021506
      0.059631
      0.068242
    
    
      785260
      0.044185
      0.258607
      0.053640
      0.059585
      0.363492
      0.070872
      0.021807
      0.061240
      0.066571
    
    
      785261
      0.044306
      0.255635
      0.052401
      0.059093
      0.367128
      0.071790
      0.022941
      0.058761
      0.067946
    
    
      785262
      0.044716
      0.254128
      0.052714
      0.061350
      0.366313
      0.071204
      0.022471
      0.058877
      0.068226
    
    
      785263
      0.043735
      0.253021
      0.052048
      0.060996
      0.370278
      0.070674
      0.022127
      0.061097
      0.066024
    
    
      785264
      0.042977
      0.251533
      0.052379
      0.061338
      0.370467
      0.070841
      0.022585
      0.061118
      0.066762
    
    
      785265
      0.044458
      0.251456
      0.051608
      0.059767
      0.373960
      0.068759
      0.023972
      0.060584
      0.065435
    
    
      785266
      0.043890
      0.255661
      0.049505
      0.060533
      0.372163
      0.067475
      0.023559
      0.060753
      0.066462
    
    
      785267
      0.046575
      0.251592
      0.051324
      0.062732
      0.368718
      0.067622
      0.024026
      0.061787
      0.065626
    
    
      785268
      0.044577
      0.257012
      0.052221
      0.063061
      0.369648
      0.066149
      0.022564
      0.060743
      0.064025
    
    
      785269
      0.044268
      0.254033
      0.051678
      0.064370
      0.369154
      0.066421
      0.024167
      0.060261
      0.065648
    
    
      785270
      0.048669
      0.275139
      0.050590
      0.064896
      0.351255
      0.061972
      0.024502
      0.060611
      0.062365
    
    
      785271
      0.046323
      0.276102
      0.051598
      0.061403
      0.350058
      0.065761
      0.023445
      0.063879
      0.061431
    
    
      785272
      0.047640
      0.273556
      0.053271
      0.061664
      0.350096
      0.065440
      0.023670
      0.061596
      0.063066
    
    
      785273
      0.046764
      0.272163
      0.051324
      0.059536
      0.354481
      0.066837
      0.023349
      0.060065
      0.065480
    
    
      785274
      0.047354
      0.278096
      0.051196
      0.061315
      0.347737
      0.067058
      0.023003
      0.059665
      0.064577
    
    
      785275
      0.048603
      0.280657
      0.051618
      0.058849
      0.345514
      0.067805
      0.022711
      0.059878
      0.064364
    
    
      785276
      0.046198
      0.278569
      0.054519
      0.059578
      0.346105
      0.069822
      0.021950
      0.060244
      0.063015
    
    
      785277
      0.047840
      0.276019
      0.054323
      0.060920
      0.345948
      0.068025
      0.021839
      0.060849
      0.064236
    
    
      785278
      0.046960
      0.278458
      0.054748
      0.061308
      0.344173
      0.068620
      0.021560
      0.060682
      0.063491
    
    
      785279
      0.048014
      0.276697
      0.054558
      0.059273
      0.345299
      0.069937
      0.021882
      0.059249
      0.065091
    
    
      785280
      0.047295
      0.285146
      0.055665
      0.060464
      0.336913
      0.068236
      0.021854
      0.061402
      0.063024
    
    
      785281
      0.046069
      0.283838
      0.054845
      0.057951
      0.342521
      0.068038
      0.021745
      0.060809
      0.064184
    
    
      785282
      0.047428
      0.290018
      0.054453
      0.057945
      0.336572
      0.067724
      0.021447
      0.059859
      0.064554
    
    
      785283
      0.048316
      0.289487
      0.054065
      0.058682
      0.334200
      0.070215
      0.020822
      0.059541
      0.064672
    
    
      785284
      0.048527
      0.292924
      0.056787
      0.057575
      0.331831
      0.070127
      0.021034
      0.057821
      0.063374
    
  

785285 rows × 9 columns

[plt.plot(PLDpixels[key]) for key in PLDpixels.columns.values];



In [23]:

    
spitzerData = spitzerDataRaw.copy()
for key in spitzerDataRaw.columns: 
    if key in PLDpixels.columns:
        spitzerData[key] = PLDpixels[key]



In [24]:

    
testPLD = np.array(pd.DataFrame({key:spitzerData[key] for key in spitzerData.columns.values if 'pix' in key}))
assert(not sum(abs(testPLD - np.array(PLDpixels))).all())
print('Confirmed that PLD Pixels have been Normalized to Spec')









    



Confirmed that PLD Pixels have been Normalized to Spec



In [25]:

    
notFeatures     = ['flux', 'fluxerr', 'xerr', 'yerr', 'xycov', 't_cernox']

periodMax           = spitzerData['bmjd'].values.max() - spitzerData['bmjd'].values.min()
periodMin           = np.min(np.diff(spitzerData['bmjd'].values))
spitzerData['freq'] = np.linspace(np.pi/periodMax, 4*np.pi/periodMin, spitzerData['bmjd'].values.size)

feature_columns = spitzerData.drop(notFeatures,axis=1).columns.values
features        = spitzerData.drop(notFeatures,axis=1).values
labels          = spitzerData['flux'].values



In [26]:

    
stdScaler = StandardScaler()



In [27]:

    
features_scaled = stdScaler.fit_transform(features)
labels_scaled   = labels#stdScaler.fit_transform(labels[:,None]).ravel()

idx_valtest, idx_train = train_test_split(np.arange(labels_scaled.size), test_size=0.6, random_state=42)
idx_val, idx_test      = train_test_split(idx_valtest                  , test_size=0.5, random_state=42)

x_val   = features_scaled[idx_val]
x_test  = features_scaled[idx_test]
x_train = features_scaled[idx_train]

y_val   = labels_scaled[idx_val]
y_test  = labels_scaled[idx_test]
y_train = labels_scaled[idx_train]

y_val_err   = spitzerData['fluxerr'].values[idx_val]
y_test_err  = spitzerData['fluxerr'].values[idx_test]
y_train_err = spitzerData['fluxerr'].values[idx_train]

print(x_val.shape  , 'validation samples')
print(x_train.shape, 'train samples')
print(x_test.shape , 'test samples')









    



(157057, 19) validation samples
(471171, 19) train samples
(157057, 19) test samples



In [28]:

    
train_df    = pd.DataFrame(np.c_[x_train, y_train], columns=list(feature_columns) + ['flux'])
test_df     = pd.DataFrame(np.c_[x_test , y_test ], columns=list(feature_columns) + ['flux'])
evaluate_df = pd.DataFrame(np.c_[x_val  , y_val  ], columns=list(feature_columns) + ['flux'])

We only take first 1000 rows for training/testing and last 500 row for evaluation.

This done so that this script does not consume a lot of kaggle system resources.



In [29]:

    
# train_df = df_train_ori.head(1000)
# evaluate_df = df_train_ori.tail(500)

# test_df = df_test_ori.head(1000)

# MODEL_DIR = "tf_model_spitzer/withNormalization_drop50/relu"
# MODEL_DIR = "tf_model_spitzer/adamOptimizer_with_drop50/relu"
MODEL_DIR = "tf_model_spitzer/adamOptimizer/drop50/elu/"

print("train_df.shape = "   , train_df.shape)
print("test_df.shape = "    , test_df.shape)
print("evaluate_df.shape = ", evaluate_df.shape)









    



train_df.shape =  (471171, 20)
test_df.shape =  (157057, 20)
evaluate_df.shape =  (157057, 20)

Filtering Categorical and Continuous features

We store Categorical, Continuous and Target features names in different variables. This will be helpful in later steps.



In [30]:

    
# categorical_features = [feature for feature in features if 'cat' in feature]
categorical_features  = []
continuous_features   = [feature for feature in train_df.columns]# if 'cat' in feature]
LABEL_COLUMN          = 'flux'

Converting Data into Tensors

When building a TF.Learn model, the input data is specified by means of an Input Builder function. This builder function will not be called until it is later passed to TF.Learn methods such as fit and evaluate. The purpose of this function is to construct the input data, which is represented in the form of Tensors or SparseTensors.

Note that input_fn will be called while constructing the TensorFlow graph, not while running the graph. What it is returning is a representation of the input data as the fundamental unit of TensorFlow computations, a Tensor (or SparseTensor).

More detail on input_fn.



In [31]:

    
# Converting Data into Tensors
def input_fn(df, training = True):
    # Creates a dictionary mapping from each continuous feature column name (k) to
    # the values of that column stored in a constant Tensor.
    continuous_cols = {k: tf.constant(df[k].values)
                       for k in continuous_features}

    # Creates a dictionary mapping from each categorical feature column name (k)
    # to the values of that column stored in a tf.SparseTensor.
    # categorical_cols = {k: tf.SparseTensor(
    #     indices=[[i, 0] for i in range(df[k].size)],
    #     values=df[k].values,
    #     shape=[df[k].size, 1])
    #     for k in categorical_features}

    # Merges the two dictionaries into one.
    feature_cols = continuous_cols
    # feature_cols = dict(list(continuous_cols.items()) + list(categorical_cols.items()))
    
    if training:
        # Converts the label column into a constant Tensor.
        label = tf.constant(df[LABEL_COLUMN].values)

        # Returns the feature columns and the label.
        return feature_cols, label
    
    # Returns the feature columns    
    return feature_cols

def train_input_fn():
    return input_fn(train_df, training=True)

def eval_input_fn():
    return input_fn(evaluate_df, training=False)

def test_input_fn():
    return input_fn(test_df, training=False)

Selecting and Engineering Features for the Model

We use tf.learn's concept of FeatureColumn which help in transforming raw data into suitable input features.

These engineered features will be used when we construct our model.



In [32]:

    
engineered_features = []

for continuous_feature in continuous_features:
    engineered_features.append(
        tf.contrib.layers.real_valued_column(continuous_feature))

Defining The Regression Model

Following is the simple DNNRegressor model. More detail about hidden_units, etc can be found here.

model_dir is used to save and restore our model. This is because once we have trained the model we don't want to train it again, if we only want to predict on new data-set.



In [33]:

    
config = tf.contrib.learn.RunConfig(tf_random_seed=42) # not shown in the config



In [34]:

    
n_inputs  = features_scaled.shape[0]
n_features= features_scaled.shape[1]



In [35]:

    
n_inputs  = x_train.shape[0]
n_features= x_train.shape[1]
n_inputs, n_features









    Out[35]:





(471171, 19)



In [54]:

    
n_hidden1 = n_features
n_hidden2 = n_features
n_hidden3 = n_features
n_outputs = n_inputs   # because: regression



In [55]:

    
reset_graph()



In [56]:

    
with tf.name_scope('data'):
    X   = tf.placeholder(tf.float32, shape=(None, n_features), name="X")
    y   = tf.placeholder(tf.float32, shape=(None), name="y")
    unc = tf.placeholder(tf.float32, shape=(None), name="unc")



In [57]:

    
training = tf.placeholder_with_default(False, shape=(), name='training')

l1_reg       = 0.001
l2_reg       = 0.0001
dropout_rate = 0.5  # == 1 - keep_prob

X_bnorm= tf.layers.batch_normalization(X)
X_drop = tf.layers.dropout(X_bnorm, dropout_rate, training=training)

with tf.name_scope("dnn"):
    he_init      = tf.contrib.layers.variance_scaling_initializer()
    
    hidden1      = tf.layers.dense(X_drop , n_hidden1, name="hidden1", 
                        activation=tf.nn.elu, kernel_initializer=he_init,
                        kernel_regularizer=tf.contrib.layers.l1_l2_regularizer(scale_l1=l1_reg, scale_l2=l2_reg))
    
    hidden1_bnorm= tf.layers.batch_normalization(hidden1, training=training, momentum=0.9)
    hidden1_drop = tf.layers.dropout(hidden1_bnorm, dropout_rate, training=training)
    
    hidden2      = tf.layers.dense(hidden1_drop, n_hidden2, name="hidden2", 
                        activation=tf.nn.elu, kernel_initializer=he_init,
                        kernel_regularizer=tf.contrib.layers.l1_l2_regularizer(scale_l1=l1_reg, scale_l2=l2_reg))
    
    hidden2_bnorm= tf.layers.batch_normalization(hidden2, training=training, momentum=0.9)
    hidden2_drop = tf.layers.dropout(hidden2_bnorm, dropout_rate, training=training)
    
    hidden3      = tf.layers.dense(hidden2_drop, n_hidden3, name="hidden3", 
                        activation=tf.nn.elu, kernel_initializer=he_init,
                        kernel_regularizer=tf.contrib.layers.l1_l2_regularizer(scale_l1=l1_reg, scale_l2=l2_reg))
    
    hidden3_bnorm= tf.layers.batch_normalization(hidden3, training=training, momentum=0.9)
    hidden3_drop = tf.layers.dropout(hidden3_bnorm, dropout_rate, training=training)
    
    output  = tf.layers.dense(hidden3_drop, n_outputs, name="outputs")



In [58]:

    
with tf.name_scope("loss"):
    def tf_nll(labels, output, uncs, coeff=1):
        error = output - labels
        return tf.reduce_sum(tf.divide(tf.squared_difference(output, labels) , tf.square(uncs)))# + tf.log(tf.square(uncs))
        #return tf.reduce_sum(1 * (coeff * np.log(2*np.pi) + coeff * tf.log(uncs) + (0.5/uncs) * tf.pow(error, 2)))
    
    negloglike  = tf_nll(labels=y, output=output, uncs=unc)
    
    reg_losses  = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
    loss        = tf.add_n([negloglike] + reg_losses, name="chisq")



In [59]:

    
with tf.name_scope("eval"):
    accuracy  = tf.reduce_mean(tf.squared_difference(output, y, name="accuracy"))
    
    SqErrRatio= tf.divide(accuracy,  tf.reduce_mean(tf.squared_difference(y, tf.reduce_mean(y))))
    r2_acc    = 1.0 - SqErrRatio
    
    chsiqMean = tf_nll(labels=y, output=tf.reduce_mean(y), uncs=unc)
    chisqModel= tf_nll(labels=y, output=output, uncs=unc)
    rho2_acc  = 1.0 - chisqModel / chsiqMean



In [60]:

    
learning_rate = 0.001

# with tf.name_scope("train"):
optimizer = tf.train.AdamOptimizer(learning_rate)
training_op = optimizer.minimize(loss)

init = tf.global_variables_initializer()



In [61]:

    
from datetime import datetime

now = datetime.utcnow().strftime("%Y%m%d%H%M%S")
root_logdir = "tf_logs"
logdir = "{}/run-{}/".format(root_logdir, now)

print(logdir)









    



tf_logs/run-20171204202811/



In [62]:

    
mse_summary  = tf.summary.scalar('train_acc' , accuracy  )
loss_summary = tf.summary.scalar('loss'      , loss      )
nll_summary  = tf.summary.scalar('negloglike', negloglike)
r2s_summary  = tf.summary.scalar('r2_acc'    , r2_acc    )
p2s_summary  = tf.summary.scalar('rho2_acc'  , rho2_acc  )
val_summary  = tf.summary.scalar('val_acc'   , accuracy  )

# hid1_hist    = tf.summary.histogram('hidden1', hidden1)
# hid2_hist    = tf.summary.histogram('hidden1', hidden1)
# hid3_hist    = tf.summary.histogram('hidden1', hidden1)

file_writer = tf.summary.FileWriter(logdir, tf.get_default_graph())



In [63]:

    
init = tf.global_variables_initializer()
saver = tf.train.Saver()



In [64]:

    
batch_size = 50
n_epochs   = 10
n_batches  = n_outputs // batch_size



In [65]:

    
def fetch_batch(epoch, batch_index, batch_size, trainingNow=True):
    np.random.seed(epoch * n_batches + batch_index)
    indices = np.random.randint(y_train.size, size=batch_size)
    X_batch = x_train[indices]
    y_batch = y_train.reshape(-1, 1)[indices]
    u_batch = y_train_err.reshape(-1, 1)[indices]
    
    return {training: trainingNow, X: X_batch, y: y_batch, unc:u_batch}



In [66]:

    
def fetch_acc_batch(epoch, batch_index, batch_size, batch_size_mod=10, trainingNow=False):
    np.random.seed(epoch * n_batches + batch_index)
    indices = np.random.randint(y_val.size, size=batch_size_mod*batch_size)
    X_batch = x_val[indices]
    y_batch = y_val.reshape(-1, 1)[indices]
    u_batch = y_val_err.reshape(-1, 1)[indices]
    return {training: trainingNow, X: X_batch, y: y_batch, unc:u_batch}

from datetime import datetime def log_dir(prefix=""): now = datetime.utcnow().strftime("%Y%m%d%H%M%S") root_logdir = "tf_logs" if prefix: prefix += "-" name = prefix + "run-" + now return "{}/{}/".format(root_logdir, name)

file_writer = tf.summary.FileWriter("logs/spitzer", tf.get_default_graph())



In [67]:

    
def write_to_tensorboard(epoch, batch_index, feed_dict_here):
    step = epoch * n_batches + batch_index
    
    feed_dict_here[training] = False
    
    list_of_summaries = [nll_summary, mse_summary, loss_summary, r2s_summary, p2s_summary]
    
    for summary in list_of_summaries:
        file_writer.add_summary(summary=summary.eval(feed_dict=feed_dict_now), global_step=step)
    
    acc_feed_dict = fetch_acc_batch(epoch, batch_index, batch_size)
    file_writer.add_summary(summary=val_summary.eval(feed_dict=acc_feed_dict), global_step=step)
    
    del feed_dict_here, acc_feed_dict



In [68]:

    
def print_update(feed_dict_now, epoch, batch_index, batch_size, n_batches, batch_size_mod=10):
    
    acc_test  = accuracy.eval(feed_dict=fetch_acc_batch(epoch, batch_index, batch_size))
    acc_train = accuracy.eval(feed_dict=feed_dict_now)
    
    r2s_test  = r2_acc.eval(feed_dict=fetch_acc_batch(epoch, batch_index, batch_size))
    rho2_test = rho2_acc.eval(feed_dict=fetch_acc_batch(epoch, batch_index, batch_size))
    
    step = epoch * n_batches + batch_index
    
    print(epoch, step, "Train accuracy:"  , acc_train, 
                       "Test accuracy:"   , acc_test , 
                       "R2 Test Score:"   , r2s_test , 
                       "Rho2 Test Score:" , rho2_test)



In [ ]:

    
batch_write_skip = 10

with tf.Session() as sess:
    sess.run(init)
    for epoch in tqdm_notebook(range(n_epochs), total=n_epochs, desc='epoch'):
        for batch_index in tqdm_notebook(range(n_batches), total=n_batches, desc='batch'):
            feed_dict_now = fetch_batch(epoch, batch_index, batch_size)
            
            if batch_index % batch_write_skip == 0 and batch_index > 0: 
                write_to_tensorboard(epoch, batch_index, feed_dict_now)
                print_update(feed_dict_now, epoch, batch_index, batch_size, n_batches)
            
            sess.run(training_op, feed_dict=feed_dict_now)
        
        print_update(feed_dict_now, epoch, batch_index, batch_size, n_batches, batch_size_mod=100)
    
    save_path = saver.save(sess, MODEL_DIR + "/spitzer_calibration_20_20_20_final.ckpt")









    



0 10 Train accuracy: 0.158532 Test accuracy: 0.071215 R2 Test Score: -11006.3 Rho2 Test Score: -1.12962e+10
0 20 Train accuracy: 0.125567 Test accuracy: 0.071215 R2 Test Score: -10273.9 Rho2 Test Score: -8.57222e+09
0 30 Train accuracy: 0.104778 Test accuracy: 0.0553633 R2 Test Score: -8635.42 Rho2 Test Score: -7.6576e+09
0 40 Train accuracy: 0.0805021 Test accuracy: 0.0399221 R2 Test Score: -6530.83 Rho2 Test Score: -6.62111e+09
0 50 Train accuracy: 0.0745774 Test accuracy: 0.0364796 R2 Test Score: -5909.11 Rho2 Test Score: -5.39523e+09
0 60 Train accuracy: 0.0450062 Test accuracy: 0.0359647 R2 Test Score: -5524.67 Rho2 Test Score: -4.10585e+09
0 70 Train accuracy: 0.0422732 Test accuracy: 0.0356075 R2 Test Score: -5338.62 Rho2 Test Score: -3.28319e+09
0 80 Train accuracy: 0.0359875 Test accuracy: 0.0237533 R2 Test Score: -3746.3 Rho2 Test Score: -2.57054e+09
0 90 Train accuracy: 0.0305337 Test accuracy: 0.0188536 R2 Test Score: -2970.85 Rho2 Test Score: -2.10722e+09
0 100 Train accuracy: 0.0218943 Test accuracy: 0.0167642 R2 Test Score: -2626.54 Rho2 Test Score: -1.50412e+09
0 110 Train accuracy: 0.0185946 Test accuracy: 0.010654 R2 Test Score: -1734.19 Rho2 Test Score: -1.12446e+09
0 120 Train accuracy: 0.0114023 Test accuracy: 0.00890187 R2 Test Score: -1531.34 Rho2 Test Score: -8.67392e+08
0 130 Train accuracy: 0.00707138 Test accuracy: 0.00682 R2 Test Score: -1035.06 Rho2 Test Score: -6.10935e+08
0 140 Train accuracy: 0.00501125 Test accuracy: 0.00555864 R2 Test Score: -783.547 Rho2 Test Score: -4.4361e+08
0 150 Train accuracy: 0.00385177 Test accuracy: 0.00395655 R2 Test Score: -565.878 Rho2 Test Score: -3.02623e+08
0 160 Train accuracy: 0.0036507 Test accuracy: 0.0039289 R2 Test Score: -555.528 Rho2 Test Score: -3.05151e+08
0 170 Train accuracy: 0.00322754 Test accuracy: 0.00370746 R2 Test Score: -567.917 Rho2 Test Score: -3.19656e+08
0 180 Train accuracy: 0.00578816 Test accuracy: 0.00342156 R2 Test Score: -497.099 Rho2 Test Score: -2.79458e+08
0 190 Train accuracy: 0.00207758 Test accuracy: 0.00199758 R2 Test Score: -297.574 Rho2 Test Score: -1.75727e+08
0 200 Train accuracy: 0.00210587 Test accuracy: 0.00164398 R2 Test Score: -226.467 Rho2 Test Score: -1.26951e+08
0 210 Train accuracy: 0.00194184 Test accuracy: 0.00173811 R2 Test Score: -265.578 Rho2 Test Score: -1.54797e+08
0 220 Train accuracy: 0.00134968 Test accuracy: 0.00140933 R2 Test Score: -213.915 Rho2 Test Score: -1.21805e+08
0 230 Train accuracy: 0.00212708 Test accuracy: 0.00111273 R2 Test Score: -202.957 Rho2 Test Score: -1.1154e+08
0 240 Train accuracy: 0.000834495 Test accuracy: 0.00106446 R2 Test Score: -160.194 Rho2 Test Score: -8.82165e+07
0 250 Train accuracy: 0.000888972 Test accuracy: 0.00102365 R2 Test Score: -151.721 Rho2 Test Score: -9.0048e+07
0 260 Train accuracy: 0.00186162 Test accuracy: 0.0014718 R2 Test Score: -243.525 Rho2 Test Score: -1.37735e+08
0 270 Train accuracy: 0.00272516 Test accuracy: 0.00158431 R2 Test Score: -269.649 Rho2 Test Score: -1.65758e+08
0 280 Train accuracy: 0.000947861 Test accuracy: 0.00130023 R2 Test Score: -206.547 Rho2 Test Score: -1.24579e+08
0 290 Train accuracy: 0.0004785 Test accuracy: 0.000571992 R2 Test Score: -80.0382 Rho2 Test Score: -4.51518e+07
0 300 Train accuracy: 0.000693604 Test accuracy: 0.000710321 R2 Test Score: -117.12 Rho2 Test Score: -6.48139e+07
0 310 Train accuracy: 0.00119942 Test accuracy: 0.000588017 R2 Test Score: -98.7981 Rho2 Test Score: -5.41122e+07
0 320 Train accuracy: 0.000731685 Test accuracy: 0.000480503 R2 Test Score: -78.8902 Rho2 Test Score: -4.54304e+07
0 330 Train accuracy: 0.000749659 Test accuracy: 0.000804681 R2 Test Score: -130.833 Rho2 Test Score: -7.48925e+07
0 340 Train accuracy: 0.000969955 Test accuracy: 0.000704501 R2 Test Score: -114.424 Rho2 Test Score: -6.66612e+07
0 350 Train accuracy: 0.000707522 Test accuracy: 0.000467309 R2 Test Score: -74.3377 Rho2 Test Score: -4.41734e+07
0 360 Train accuracy: 0.000545374 Test accuracy: 0.000597141 R2 Test Score: -88.214 Rho2 Test Score: -4.86139e+07
0 370 Train accuracy: 0.000342176 Test accuracy: 0.0003844 R2 Test Score: -58.8351 Rho2 Test Score: -3.37737e+07
0 380 Train accuracy: 0.000648558 Test accuracy: 0.000309433 R2 Test Score: -48.7075 Rho2 Test Score: -2.79598e+07
0 390 Train accuracy: 0.000626296 Test accuracy: 0.000427757 R2 Test Score: -64.534 Rho2 Test Score: -3.82095e+07
0 400 Train accuracy: 0.000791524 Test accuracy: 0.000336804 R2 Test Score: -50.8782 Rho2 Test Score: -2.98781e+07
0 410 Train accuracy: 0.000392163 Test accuracy: 0.00055141 R2 Test Score: -84.6187 Rho2 Test Score: -4.74499e+07
0 420 Train accuracy: 0.000647285 Test accuracy: 0.000556367 R2 Test Score: -80.2738 Rho2 Test Score: -4.70017e+07
0 430 Train accuracy: 0.000520115 Test accuracy: 0.000415893 R2 Test Score: -55.6551 Rho2 Test Score: -3.34822e+07
0 440 Train accuracy: 0.000536631 Test accuracy: 0.000307746 R2 Test Score: -45.5666 Rho2 Test Score: -2.62724e+07
0 450 Train accuracy: 0.000265391 Test accuracy: 0.000325695 R2 Test Score: -51.9997 Rho2 Test Score: -2.89553e+07
0 460 Train accuracy: 0.000395462 Test accuracy: 0.000334491 R2 Test Score: -55.858 Rho2 Test Score: -3.35438e+07
0 470 Train accuracy: 0.000320799 Test accuracy: 0.000215309 R2 Test Score: -36.3898 Rho2 Test Score: -2.2335e+07
0 480 Train accuracy: 0.000410502 Test accuracy: 0.000248747 R2 Test Score: -40.4551 Rho2 Test Score: -2.25735e+07
0 490 Train accuracy: 0.000367137 Test accuracy: 0.000307608 R2 Test Score: -51.2898 Rho2 Test Score: -2.93633e+07
0 500 Train accuracy: 0.000334478 Test accuracy: 0.000300516 R2 Test Score: -45.0384 Rho2 Test Score: -2.48768e+07
0 510 Train accuracy: 0.000321337 Test accuracy: 0.000261292 R2 Test Score: -47.4421 Rho2 Test Score: -2.68051e+07
0 520 Train accuracy: 0.000326694 Test accuracy: 0.000289902 R2 Test Score: -45.4823 Rho2 Test Score: -2.59881e+07
0 530 Train accuracy: 0.000247992 Test accuracy: 0.00024454 R2 Test Score: -38.6343 Rho2 Test Score: -2.12661e+07
0 540 Train accuracy: 0.000466036 Test accuracy: 0.00024627 R2 Test Score: -36.2947 Rho2 Test Score: -2.08878e+07
0 550 Train accuracy: 0.000155841 Test accuracy: 0.000170384 R2 Test Score: -26.8 Rho2 Test Score: -1.52293e+07
0 560 Train accuracy: 0.000123145 Test accuracy: 0.000172691 R2 Test Score: -24.4884 Rho2 Test Score: -1.35503e+07
0 570 Train accuracy: 0.000347729 Test accuracy: 0.000174749 R2 Test Score: -25.906 Rho2 Test Score: -1.51599e+07
0 580 Train accuracy: 0.000186283 Test accuracy: 0.000116824 R2 Test Score: -16.7389 Rho2 Test Score: -9.51404e+06
0 590 Train accuracy: 0.000145892 Test accuracy: 9.32115e-05 R2 Test Score: -14.5301 Rho2 Test Score: -9.00862e+06
0 600 Train accuracy: 9.67503e-05 Test accuracy: 0.000107632 R2 Test Score: -18.001 Rho2 Test Score: -1.09157e+07
0 610 Train accuracy: 0.00014456 Test accuracy: 0.000107262 R2 Test Score: -16.5058 Rho2 Test Score: -9.61606e+06
0 620 Train accuracy: 0.000189123 Test accuracy: 0.00017541 R2 Test Score: -23.2534 Rho2 Test Score: -1.32888e+07
0 630 Train accuracy: 0.000273532 Test accuracy: 0.000185967 R2 Test Score: -24.7435 Rho2 Test Score: -1.43962e+07
0 640 Train accuracy: 0.000148712 Test accuracy: 0.000152493 R2 Test Score: -24.3274 Rho2 Test Score: -1.42216e+07
0 650 Train accuracy: 0.000118657 Test accuracy: 0.000136459 R2 Test Score: -21.215 Rho2 Test Score: -1.21103e+07
0 660 Train accuracy: 0.000150302 Test accuracy: 0.000124267 R2 Test Score: -19.5157 Rho2 Test Score: -1.12723e+07
0 670 Train accuracy: 0.000231799 Test accuracy: 0.000130404 R2 Test Score: -19.2953 Rho2 Test Score: -1.10901e+07
0 680 Train accuracy: 0.000256654 Test accuracy: 0.000149269 R2 Test Score: -20.2722 Rho2 Test Score: -1.26465e+07
0 690 Train accuracy: 0.000273426 Test accuracy: 0.000150383 R2 Test Score: -22.6831 Rho2 Test Score: -1.41136e+07
0 700 Train accuracy: 0.000133432 Test accuracy: 8.56543e-05 R2 Test Score: -11.4594 Rho2 Test Score: -6.97448e+06
0 710 Train accuracy: 6.63322e-05 Test accuracy: 9.52739e-05 R2 Test Score: -15.1466 Rho2 Test Score: -9.21522e+06
0 720 Train accuracy: 0.000124668 Test accuracy: 0.000121589 R2 Test Score: -15.5859 Rho2 Test Score: -8.81075e+06
0 730 Train accuracy: 8.3341e-05 Test accuracy: 9.20068e-05 R2 Test Score: -14.0609 Rho2 Test Score: -8.93091e+06
0 740 Train accuracy: 0.000103088 Test accuracy: 8.41943e-05 R2 Test Score: -12.0586 Rho2 Test Score: -7.07985e+06
0 750 Train accuracy: 0.000114712 Test accuracy: 7.37204e-05 R2 Test Score: -10.1301 Rho2 Test Score: -5.95839e+06
0 760 Train accuracy: 7.00287e-05 Test accuracy: 8.61474e-05 R2 Test Score: -12.5619 Rho2 Test Score: -7.69009e+06
0 770 Train accuracy: 0.000267834 Test accuracy: 0.00011662 R2 Test Score: -19.5516 Rho2 Test Score: -1.12056e+07
0 780 Train accuracy: 7.7146e-05 Test accuracy: 0.000105878 R2 Test Score: -14.348 Rho2 Test Score: -8.57922e+06
0 790 Train accuracy: 7.82714e-05 Test accuracy: 0.000135762 R2 Test Score: -19.5215 Rho2 Test Score: -1.12351e+07
0 800 Train accuracy: 0.000117453 Test accuracy: 9.51766e-05 R2 Test Score: -15.3004 Rho2 Test Score: -8.9079e+06
0 810 Train accuracy: 9.90575e-05 Test accuracy: 6.61249e-05 R2 Test Score: -9.76602 Rho2 Test Score: -6.10241e+06
0 820 Train accuracy: 8.99323e-05 Test accuracy: 5.49329e-05 R2 Test Score: -8.32516 Rho2 Test Score: -5.58057e+06
0 830 Train accuracy: 4.73812e-05 Test accuracy: 4.60622e-05 R2 Test Score: -5.77326 Rho2 Test Score: -3.81108e+06
0 840 Train accuracy: 5.51418e-05 Test accuracy: 5.61535e-05 R2 Test Score: -7.46502 Rho2 Test Score: -4.4552e+06
0 850 Train accuracy: 3.97006e-05 Test accuracy: 5.03979e-05 R2 Test Score: -6.76821 Rho2 Test Score: -4.5508e+06
0 860 Train accuracy: 6.63494e-05 Test accuracy: 4.82276e-05 R2 Test Score: -6.91728 Rho2 Test Score: -5.02067e+06
0 870 Train accuracy: 6.00271e-05 Test accuracy: 5.05677e-05 R2 Test Score: -6.78765 Rho2 Test Score: -4.50372e+06
0 880 Train accuracy: 5.07523e-05 Test accuracy: 7.30026e-05 R2 Test Score: -9.87765 Rho2 Test Score: -5.90736e+06
0 890 Train accuracy: 5.31142e-05 Test accuracy: 6.56539e-05 R2 Test Score: -9.25025 Rho2 Test Score: -5.61467e+06
0 900 Train accuracy: 4.94833e-05 Test accuracy: 4.8646e-05 R2 Test Score: -6.52991 Rho2 Test Score: -4.38805e+06
0 910 Train accuracy: 9.65668e-05 Test accuracy: 3.79976e-05 R2 Test Score: -5.25429 Rho2 Test Score: -3.53753e+06
0 920 Train accuracy: 0.000119716 Test accuracy: 4.84824e-05 R2 Test Score: -6.02312 Rho2 Test Score: -4.22225e+06
0 930 Train accuracy: 5.58674e-05 Test accuracy: 6.79761e-05 R2 Test Score: -11.2708 Rho2 Test Score: -6.71185e+06
0 940 Train accuracy: 9.98835e-05 Test accuracy: 5.72219e-05 R2 Test Score: -7.52285 Rho2 Test Score: -4.73994e+06
0 950 Train accuracy: 7.775e-05 Test accuracy: 4.46411e-05 R2 Test Score: -5.56421 Rho2 Test Score: -3.64118e+06
0 960 Train accuracy: 5.27789e-05 Test accuracy: 4.3532e-05 R2 Test Score: -5.48614 Rho2 Test Score: -3.58424e+06
0 970 Train accuracy: 4.94512e-05 Test accuracy: 3.84824e-05 R2 Test Score: -5.0052 Rho2 Test Score: -3.33771e+06
0 980 Train accuracy: 3.71423e-05 Test accuracy: 3.92498e-05 R2 Test Score: -5.58631 Rho2 Test Score: -3.65298e+06
0 990 Train accuracy: 4.58747e-05 Test accuracy: 4.0144e-05 R2 Test Score: -4.58434 Rho2 Test Score: -3.12974e+06
0 1000 Train accuracy: 3.89518e-05 Test accuracy: 3.83116e-05 R2 Test Score: -5.58726 Rho2 Test Score: -3.671e+06
0 1010 Train accuracy: 5.15532e-05 Test accuracy: 5.73903e-05 R2 Test Score: -7.60598 Rho2 Test Score: -4.86315e+06
0 1020 Train accuracy: 0.000110197 Test accuracy: 8.0638e-05 R2 Test Score: -10.6219 Rho2 Test Score: -6.4689e+06
0 1030 Train accuracy: 6.19854e-05 Test accuracy: 8.68068e-05 R2 Test Score: -13.8582 Rho2 Test Score: -8.40652e+06
0 1040 Train accuracy: 4.8349e-05 Test accuracy: 5.67667e-05 R2 Test Score: -7.3382 Rho2 Test Score: -4.70426e+06
0 1050 Train accuracy: 5.75649e-05 Test accuracy: 3.91664e-05 R2 Test Score: -5.92353 Rho2 Test Score: -3.70913e+06
0 1060 Train accuracy: 5.02513e-05 Test accuracy: 5.11247e-05 R2 Test Score: -7.76407 Rho2 Test Score: -5.12089e+06
0 1070 Train accuracy: 7.36699e-05 Test accuracy: 4.86093e-05 R2 Test Score: -7.01073 Rho2 Test Score: -4.67767e+06
0 1080 Train accuracy: 2.7298e-05 Test accuracy: 3.00124e-05 R2 Test Score: -4.01874 Rho2 Test Score: -2.76548e+06
0 1090 Train accuracy: 4.45222e-05 Test accuracy: 3.24806e-05 R2 Test Score: -4.09272 Rho2 Test Score: -2.96736e+06
0 1100 Train accuracy: 5.77205e-05 Test accuracy: 3.26321e-05 R2 Test Score: -3.67897 Rho2 Test Score: -2.64787e+06
0 1110 Train accuracy: 2.17369e-05 Test accuracy: 2.42403e-05 R2 Test Score: -2.72861 Rho2 Test Score: -2.08817e+06
0 1120 Train accuracy: 1.96154e-05 Test accuracy: 2.40903e-05 R2 Test Score: -2.56798 Rho2 Test Score: -2.02568e+06
0 1130 Train accuracy: 4.84616e-05 Test accuracy: 3.3011e-05 R2 Test Score: -4.09283 Rho2 Test Score: -2.83663e+06
0 1140 Train accuracy: 2.8495e-05 Test accuracy: 1.97749e-05 R2 Test Score: -2.29942 Rho2 Test Score: -1.87991e+06
0 1150 Train accuracy: 2.51324e-05 Test accuracy: 2.12906e-05 R2 Test Score: -2.23117 Rho2 Test Score: -1.82288e+06
0 1160 Train accuracy: 6.65745e-05 Test accuracy: 4.84448e-05 R2 Test Score: -6.82899 Rho2 Test Score: -4.49639e+06
0 1170 Train accuracy: 3.84436e-05 Test accuracy: 3.53161e-05 R2 Test Score: -4.18279 Rho2 Test Score: -2.88061e+06
0 1180 Train accuracy: 3.78484e-05 Test accuracy: 3.71856e-05 R2 Test Score: -5.3765 Rho2 Test Score: -3.4536e+06
0 1190 Train accuracy: 3.59937e-05 Test accuracy: 2.02107e-05 R2 Test Score: -2.39249 Rho2 Test Score: -1.91545e+06
0 1200 Train accuracy: 2.87818e-05 Test accuracy: 1.94881e-05 R2 Test Score: -2.17238 Rho2 Test Score: -1.75714e+06
0 1210 Train accuracy: 2.28225e-05 Test accuracy: 2.06208e-05 R2 Test Score: -2.1101 Rho2 Test Score: -1.75272e+06
0 1220 Train accuracy: 1.99637e-05 Test accuracy: 2.51202e-05 R2 Test Score: -3.07958 Rho2 Test Score: -2.27384e+06
0 1230 Train accuracy: 1.86731e-05 Test accuracy: 2.02506e-05 R2 Test Score: -2.39426 Rho2 Test Score: -1.87455e+06
0 1240 Train accuracy: 1.80743e-05 Test accuracy: 1.84809e-05 R2 Test Score: -1.69172 Rho2 Test Score: -1.51234e+06
0 1250 Train accuracy: 9.04709e-06 Test accuracy: 1.82154e-05 R2 Test Score: -1.57838 Rho2 Test Score: -1.48012e+06
0 1260 Train accuracy: 2.382e-05 Test accuracy: 2.51915e-05 R2 Test Score: -2.71516 Rho2 Test Score: -2.10786e+06
0 1270 Train accuracy: 2.01392e-05 Test accuracy: 1.73865e-05 R2 Test Score: -1.49358 Rho2 Test Score: -1.37116e+06
0 1280 Train accuracy: 3.86399e-05 Test accuracy: 2.60187e-05 R2 Test Score: -3.14207 Rho2 Test Score: -2.61957e+06
0 1290 Train accuracy: 1.7573e-05 Test accuracy: 1.39159e-05 R2 Test Score: -1.25018 Rho2 Test Score: -1.30704e+06
0 1300 Train accuracy: 2.86358e-05 Test accuracy: 1.77532e-05 R2 Test Score: -1.66722 Rho2 Test Score: -1.46674e+06
0 1310 Train accuracy: 1.74579e-05 Test accuracy: 1.73865e-05 R2 Test Score: -1.86847 Rho2 Test Score: -1.55216e+06
0 1320 Train accuracy: 2.41781e-05 Test accuracy: 2.62365e-05 R2 Test Score: -3.10134 Rho2 Test Score: -2.33225e+06
0 1330 Train accuracy: 5.41225e-05 Test accuracy: 2.99098e-05 R2 Test Score: -3.68256 Rho2 Test Score: -2.58085e+06
0 1340 Train accuracy: 2.64559e-05 Test accuracy: 2.13128e-05 R2 Test Score: -2.53405 Rho2 Test Score: -2.12079e+06
0 1350 Train accuracy: 1.44046e-05 Test accuracy: 1.5242e-05 R2 Test Score: -1.41704 Rho2 Test Score: -1.35999e+06
0 1360 Train accuracy: 9.78413e-06 Test accuracy: 1.3173e-05 R2 Test Score: -0.928299 Rho2 Test Score: -1.12944e+06
0 1370 Train accuracy: 1.59116e-05 Test accuracy: 1.13788e-05 R2 Test Score: -0.89394 Rho2 Test Score: -1.12855e+06
0 1380 Train accuracy: 1.55798e-05 Test accuracy: 1.30511e-05 R2 Test Score: -0.847969 Rho2 Test Score: -1.05251e+06
0 1390 Train accuracy: 1.78094e-05 Test accuracy: 1.78009e-05 R2 Test Score: -1.65337 Rho2 Test Score: -1.44916e+06
0 1400 Train accuracy: 1.58819e-05 Test accuracy: 1.32317e-05 R2 Test Score: -1.13823 Rho2 Test Score: -1.22158e+06
0 1410 Train accuracy: 7.80607e-06 Test accuracy: 9.99078e-06 R2 Test Score: -0.679489 Rho2 Test Score: -937847.0
0 1420 Train accuracy: 1.41259e-05 Test accuracy: 1.62298e-05 R2 Test Score: -1.98264 Rho2 Test Score: -1.67529e+06
0 1430 Train accuracy: 1.64398e-05 Test accuracy: 1.51905e-05 R2 Test Score: -1.28332 Rho2 Test Score: -1.26833e+06
0 1440 Train accuracy: 2.91841e-05 Test accuracy: 1.16226e-05 R2 Test Score: -0.718319 Rho2 Test Score: -1.00687e+06
0 1450 Train accuracy: 1.80839e-05 Test accuracy: 1.37179e-05 R2 Test Score: -1.13028 Rho2 Test Score: -1.26306e+06
0 1460 Train accuracy: 1.29847e-05 Test accuracy: 1.08508e-05 R2 Test Score: -0.800512 Rho2 Test Score: -1.05406e+06
0 1470 Train accuracy: 1.39364e-05 Test accuracy: 1.20313e-05 R2 Test Score: -1.06197 Rho2 Test Score: -1.20543e+06
0 1480 Train accuracy: 6.94197e-06 Test accuracy: 8.95804e-06 R2 Test Score: -0.310616 Rho2 Test Score: -766403.0
0 1490 Train accuracy: 1.06799e-05 Test accuracy: 1.02904e-05 R2 Test Score: -0.526246 Rho2 Test Score: -878209.0
0 1500 Train accuracy: 9.38142e-06 Test accuracy: 8.93738e-06 R2 Test Score: -0.373054 Rho2 Test Score: -746031.0
0 1510 Train accuracy: 1.26066e-05 Test accuracy: 9.12041e-06 R2 Test Score: -0.436396 Rho2 Test Score: -755436.0
0 1520 Train accuracy: 1.71791e-05 Test accuracy: 1.12156e-05 R2 Test Score: -0.860253 Rho2 Test Score: -1.0697e+06
0 1530 Train accuracy: 1.37627e-05 Test accuracy: 1.04209e-05 R2 Test Score: -0.573078 Rho2 Test Score: -866296.0
0 1540 Train accuracy: 6.8185e-06 Test accuracy: 9.6698e-06 R2 Test Score: -0.477716 Rho2 Test Score: -810880.0
0 1550 Train accuracy: 1.02404e-05 Test accuracy: 9.60698e-06 R2 Test Score: -0.487 Rho2 Test Score: -830095.0
0 1560 Train accuracy: 1.33735e-05 Test accuracy: 8.69323e-06 R2 Test Score: -0.488964 Rho2 Test Score: -836138.0
0 1570 Train accuracy: 8.39914e-06 Test accuracy: 1.09825e-05 R2 Test Score: -0.935968 Rho2 Test Score: -1.09796e+06
0 1580 Train accuracy: 8.44115e-06 Test accuracy: 1.08774e-05 R2 Test Score: -0.618047 Rho2 Test Score: -929340.0
0 1590 Train accuracy: 1.72875e-05 Test accuracy: 1.09911e-05 R2 Test Score: -0.749662 Rho2 Test Score: -1.03662e+06
0 1600 Train accuracy: 9.98524e-06 Test accuracy: 9.95208e-06 R2 Test Score: -0.557729 Rho2 Test Score: -880673.0
0 1610 Train accuracy: 1.28182e-05 Test accuracy: 6.08287e-06 R2 Test Score: -0.0168619 Rho2 Test Score: -602349.0
0 1620 Train accuracy: 1.51987e-05 Test accuracy: 1.15965e-05 R2 Test Score: -0.828851 Rho2 Test Score: -1.02746e+06
0 1630 Train accuracy: 6.78703e-06 Test accuracy: 1.10469e-05 R2 Test Score: -0.690395 Rho2 Test Score: -884734.0
0 1640 Train accuracy: 1.04892e-05 Test accuracy: 8.61425e-06 R2 Test Score: -0.300583 Rho2 Test Score: -701372.0
0 1650 Train accuracy: 1.17521e-05 Test accuracy: 8.96789e-06 R2 Test Score: -0.371734 Rho2 Test Score: -774405.0
0 1660 Train accuracy: 7.74225e-06 Test accuracy: 7.57797e-06 R2 Test Score: -0.168756 Rho2 Test Score: -656834.0
0 1670 Train accuracy: 9.54749e-06 Test accuracy: 4.89593e-06 R2 Test Score: 0.183979 Rho2 Test Score: -459283.0
0 1680 Train accuracy: 1.14517e-05 Test accuracy: 6.66613e-06 R2 Test Score: 0.0559155 Rho2 Test Score: -538288.0
0 1690 Train accuracy: 6.44706e-06 Test accuracy: 6.35989e-06 R2 Test Score: -0.0141157 Rho2 Test Score: -552539.0
0 1700 Train accuracy: 1.55847e-05 Test accuracy: 9.42566e-06 R2 Test Score: -0.444291 Rho2 Test Score: -819723.0
0 1710 Train accuracy: 3.33948e-06 Test accuracy: 6.20297e-06 R2 Test Score: 0.109428 Rho2 Test Score: -545437.0
0 1720 Train accuracy: 6.09239e-06 Test accuracy: 7.56756e-06 R2 Test Score: -0.226577 Rho2 Test Score: -650934.0
0 1730 Train accuracy: 1.10697e-05 Test accuracy: 6.66594e-06 R2 Test Score: -0.174553 Rho2 Test Score: -659300.0
0 1740 Train accuracy: 1.83433e-05 Test accuracy: 6.48285e-06 R2 Test Score: -0.0241534 Rho2 Test Score: -580367.0
0 1750 Train accuracy: 7.59269e-06 Test accuracy: 6.81573e-06 R2 Test Score: -0.117302 Rho2 Test Score: -637041.0
0 1760 Train accuracy: 1.3112e-05 Test accuracy: 9.05944e-06 R2 Test Score: -0.305377 Rho2 Test Score: -762602.0
0 1770 Train accuracy: 5.21276e-06 Test accuracy: 5.65669e-06 R2 Test Score: 0.0991282 Rho2 Test Score: -513889.0
0 1780 Train accuracy: 9.87343e-06 Test accuracy: 6.66152e-06 R2 Test Score: -0.00856304 Rho2 Test Score: -562740.0
0 1790 Train accuracy: 7.11286e-06 Test accuracy: 5.93445e-06 R2 Test Score: 0.153396 Rho2 Test Score: -443505.0
0 1800 Train accuracy: 7.93761e-06 Test accuracy: 4.3483e-06 R2 Test Score: 0.256235 Rho2 Test Score: -408421.0
0 1810 Train accuracy: 7.63945e-06 Test accuracy: 5.42436e-06 R2 Test Score: 0.166002 Rho2 Test Score: -457512.0
0 1820 Train accuracy: 5.77605e-06 Test accuracy: 4.8349e-06 R2 Test Score: 0.302027 Rho2 Test Score: -390224.0
0 1830 Train accuracy: 6.37081e-06 Test accuracy: 6.43747e-06 R2 Test Score: 0.0195943 Rho2 Test Score: -561456.0
0 1840 Train accuracy: 5.68916e-06 Test accuracy: 5.75128e-06 R2 Test Score: 0.115667 Rho2 Test Score: -509481.0
0 1850 Train accuracy: 3.61083e-06 Test accuracy: 5.07904e-06 R2 Test Score: 0.192122 Rho2 Test Score: -439775.0
0 1860 Train accuracy: 4.88869e-06 Test accuracy: 4.873e-06 R2 Test Score: 0.189071 Rho2 Test Score: -451924.0
0 1870 Train accuracy: 7.97839e-06 Test accuracy: 6.47191e-06 R2 Test Score: 0.0221055 Rho2 Test Score: -546302.0
0 1880 Train accuracy: 4.10229e-06 Test accuracy: 4.51069e-06 R2 Test Score: 0.213099 Rho2 Test Score: -438065.0
0 1890 Train accuracy: 4.46107e-06 Test accuracy: 5.11933e-06 R2 Test Score: 0.174126 Rho2 Test Score: -447304.0
0 1900 Train accuracy: 3.99058e-06 Test accuracy: 4.46748e-06 R2 Test Score: 0.314026 Rho2 Test Score: -379900.0
0 1910 Train accuracy: 9.02949e-06 Test accuracy: 4.5307e-06 R2 Test Score: 0.233701 Rho2 Test Score: -443638.0
0 1920 Train accuracy: 5.44229e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.251347 Rho2 Test Score: -420240.0
0 1930 Train accuracy: 7.6147e-06 Test accuracy: 5.89197e-06 R2 Test Score: 0.0631307 Rho2 Test Score: -527734.0
0 1940 Train accuracy: 5.58593e-06 Test accuracy: 5.14007e-06 R2 Test Score: 0.240442 Rho2 Test Score: -439129.0
0 1950 Train accuracy: 5.4431e-06 Test accuracy: 4.95697e-06 R2 Test Score: 0.197942 Rho2 Test Score: -436257.0
0 1960 Train accuracy: 8.22696e-06 Test accuracy: 5.38304e-06 R2 Test Score: 0.0895784 Rho2 Test Score: -482210.0
0 1970 Train accuracy: 9.1514e-06 Test accuracy: 6.23629e-06 R2 Test Score: -0.0612725 Rho2 Test Score: -593561.0
0 1980 Train accuracy: 1.16939e-05 Test accuracy: 1.13737e-05 R2 Test Score: -0.691107 Rho2 Test Score: -883673.0
0 1990 Train accuracy: 6.62074e-06 Test accuracy: 3.96389e-06 R2 Test Score: 0.348718 Rho2 Test Score: -366924.0
0 2000 Train accuracy: 6.28718e-06 Test accuracy: 5.44564e-06 R2 Test Score: 0.178959 Rho2 Test Score: -454834.0
0 2010 Train accuracy: 4.65352e-06 Test accuracy: 4.15237e-06 R2 Test Score: 0.339645 Rho2 Test Score: -353454.0
0 2020 Train accuracy: 4.30744e-06 Test accuracy: 4.21389e-06 R2 Test Score: 0.36658 Rho2 Test Score: -365818.0
0 2030 Train accuracy: 7.00218e-06 Test accuracy: 4.59076e-06 R2 Test Score: 0.35927 Rho2 Test Score: -348819.0
0 2040 Train accuracy: 4.36046e-06 Test accuracy: 3.9056e-06 R2 Test Score: 0.378993 Rho2 Test Score: -331297.0
0 2050 Train accuracy: 4.37137e-06 Test accuracy: 5.2885e-06 R2 Test Score: 0.206474 Rho2 Test Score: -426807.0
0 2060 Train accuracy: 6.35564e-06 Test accuracy: 4.84903e-06 R2 Test Score: 0.163615 Rho2 Test Score: -462540.0
0 2070 Train accuracy: 4.29896e-06 Test accuracy: 5.07904e-06 R2 Test Score: 0.25853 Rho2 Test Score: -415579.0
0 2080 Train accuracy: 4.95429e-06 Test accuracy: 4.53639e-06 R2 Test Score: 0.3055 Rho2 Test Score: -390022.0
0 2090 Train accuracy: 5.59034e-06 Test accuracy: 4.59076e-06 R2 Test Score: 0.278913 Rho2 Test Score: -414190.0
0 2100 Train accuracy: 5.52345e-06 Test accuracy: 4.42547e-06 R2 Test Score: 0.32729 Rho2 Test Score: -382743.0
0 2110 Train accuracy: 3.70242e-06 Test accuracy: 4.52485e-06 R2 Test Score: 0.308418 Rho2 Test Score: -388642.0
0 2120 Train accuracy: 6.31337e-06 Test accuracy: 5.018e-06 R2 Test Score: 0.215741 Rho2 Test Score: -454630.0
0 2130 Train accuracy: 4.06706e-06 Test accuracy: 3.81623e-06 R2 Test Score: 0.399012 Rho2 Test Score: -330379.0
0 2140 Train accuracy: 6.94143e-06 Test accuracy: 5.46788e-06 R2 Test Score: 0.192015 Rho2 Test Score: -467223.0
0 2150 Train accuracy: 6.41396e-06 Test accuracy: 4.40574e-06 R2 Test Score: 0.32069 Rho2 Test Score: -378667.0
0 2160 Train accuracy: 5.13246e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.322617 Rho2 Test Score: -370328.0
0 2170 Train accuracy: 5.46488e-06 Test accuracy: 4.0469e-06 R2 Test Score: 0.336862 Rho2 Test Score: -372270.0
0 2180 Train accuracy: 5.75567e-06 Test accuracy: 3.92433e-06 R2 Test Score: 0.320988 Rho2 Test Score: -382579.0
0 2190 Train accuracy: 4.63195e-06 Test accuracy: 4.46869e-06 R2 Test Score: 0.32097 Rho2 Test Score: -382109.0
0 2200 Train accuracy: 4.14895e-06 Test accuracy: 4.45461e-06 R2 Test Score: 0.268143 Rho2 Test Score: -405622.0
0 2210 Train accuracy: 4.33537e-06 Test accuracy: 4.25408e-06 R2 Test Score: 0.359155 Rho2 Test Score: -345900.0
0 2220 Train accuracy: 5.78265e-06 Test accuracy: 3.5544e-06 R2 Test Score: 0.433334 Rho2 Test Score: -323438.0
0 2230 Train accuracy: 4.04614e-06 Test accuracy: 3.45917e-06 R2 Test Score: 0.444299 Rho2 Test Score: -305959.0
0 2240 Train accuracy: 4.23295e-06 Test accuracy: 3.14987e-06 R2 Test Score: 0.437238 Rho2 Test Score: -313786.0
0 2250 Train accuracy: 4.92388e-06 Test accuracy: 4.3092e-06 R2 Test Score: 0.308576 Rho2 Test Score: -380145.0
0 2260 Train accuracy: 6.49145e-06 Test accuracy: 4.07394e-06 R2 Test Score: 0.332865 Rho2 Test Score: -363602.0
0 2270 Train accuracy: 5.77296e-06 Test accuracy: 4.71283e-06 R2 Test Score: 0.315175 Rho2 Test Score: -361586.0
0 2280 Train accuracy: 3.57992e-06 Test accuracy: 4.48249e-06 R2 Test Score: 0.374334 Rho2 Test Score: -348208.0
0 2290 Train accuracy: 4.78662e-06 Test accuracy: 4.0333e-06 R2 Test Score: 0.336127 Rho2 Test Score: -362650.0
0 2300 Train accuracy: 3.45146e-06 Test accuracy: 3.36754e-06 R2 Test Score: 0.477861 Rho2 Test Score: -297619.0
0 2310 Train accuracy: 4.05123e-06 Test accuracy: 3.90731e-06 R2 Test Score: 0.363165 Rho2 Test Score: -356850.0
0 2320 Train accuracy: 4.14273e-06 Test accuracy: 4.27748e-06 R2 Test Score: 0.330078 Rho2 Test Score: -352802.0
0 2330 Train accuracy: 4.32474e-06 Test accuracy: 3.83258e-06 R2 Test Score: 0.393232 Rho2 Test Score: -328354.0
0 2340 Train accuracy: 4.55549e-06 Test accuracy: 4.13139e-06 R2 Test Score: 0.359613 Rho2 Test Score: -355794.0
0 2350 Train accuracy: 4.46464e-06 Test accuracy: 3.73399e-06 R2 Test Score: 0.338013 Rho2 Test Score: -365942.0
0 2360 Train accuracy: 5.02749e-06 Test accuracy: 4.16569e-06 R2 Test Score: 0.360766 Rho2 Test Score: -356864.0
0 2370 Train accuracy: 5.50957e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.300936 Rho2 Test Score: -380847.0
0 2380 Train accuracy: 5.32749e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.328243 Rho2 Test Score: -352903.0
0 2390 Train accuracy: 3.8915e-06 Test accuracy: 3.7455e-06 R2 Test Score: 0.404387 Rho2 Test Score: -328456.0
0 2400 Train accuracy: 5.62691e-06 Test accuracy: 4.22427e-06 R2 Test Score: 0.374198 Rho2 Test Score: -330860.0
0 2410 Train accuracy: 5.23109e-06 Test accuracy: 4.47562e-06 R2 Test Score: 0.311834 Rho2 Test Score: -361249.0
0 2420 Train accuracy: 4.76724e-06 Test accuracy: 4.00345e-06 R2 Test Score: 0.413728 Rho2 Test Score: -320088.0
0 2430 Train accuracy: 3.29594e-06 Test accuracy: 4.28984e-06 R2 Test Score: 0.384911 Rho2 Test Score: -330817.0
0 2440 Train accuracy: 4.22382e-06 Test accuracy: 3.80152e-06 R2 Test Score: 0.415171 Rho2 Test Score: -311102.0
0 2450 Train accuracy: 3.68097e-06 Test accuracy: 3.27194e-06 R2 Test Score: 0.51907 Rho2 Test Score: -284294.0
0 2460 Train accuracy: 5.02281e-06 Test accuracy: 4.46869e-06 R2 Test Score: 0.28167 Rho2 Test Score: -405158.0
0 2470 Train accuracy: 3.48104e-06 Test accuracy: 3.7607e-06 R2 Test Score: 0.3686 Rho2 Test Score: -358590.0
0 2480 Train accuracy: 5.99493e-06 Test accuracy: 3.4557e-06 R2 Test Score: 0.430811 Rho2 Test Score: -327118.0
0 2490 Train accuracy: 3.99592e-06 Test accuracy: 3.20648e-06 R2 Test Score: 0.442799 Rho2 Test Score: -315093.0
0 2500 Train accuracy: 6.91717e-06 Test accuracy: 3.85178e-06 R2 Test Score: 0.419791 Rho2 Test Score: -330884.0
0 2510 Train accuracy: 3.83543e-06 Test accuracy: 4.07504e-06 R2 Test Score: 0.370341 Rho2 Test Score: -327392.0
0 2520 Train accuracy: 3.17357e-06 Test accuracy: 3.81903e-06 R2 Test Score: 0.383827 Rho2 Test Score: -328745.0
0 2530 Train accuracy: 5.02705e-06 Test accuracy: 3.31318e-06 R2 Test Score: 0.435645 Rho2 Test Score: -315720.0
0 2540 Train accuracy: 5.1323e-06 Test accuracy: 3.44161e-06 R2 Test Score: 0.424182 Rho2 Test Score: -313138.0
0 2550 Train accuracy: 3.56581e-06 Test accuracy: 3.77954e-06 R2 Test Score: 0.419129 Rho2 Test Score: -316971.0
0 2560 Train accuracy: 5.67397e-06 Test accuracy: 3.90839e-06 R2 Test Score: 0.403602 Rho2 Test Score: -325330.0
0 2570 Train accuracy: 4.79236e-06 Test accuracy: 3.45728e-06 R2 Test Score: 0.421713 Rho2 Test Score: -308136.0
0 2580 Train accuracy: 3.35188e-06 Test accuracy: 4.0161e-06 R2 Test Score: 0.347415 Rho2 Test Score: -359420.0
0 2590 Train accuracy: 2.71311e-06 Test accuracy: 3.81309e-06 R2 Test Score: 0.360545 Rho2 Test Score: -344683.0
0 2600 Train accuracy: 4.04567e-06 Test accuracy: 4.25252e-06 R2 Test Score: 0.360396 Rho2 Test Score: -342318.0
0 2610 Train accuracy: 5.00842e-06 Test accuracy: 3.57712e-06 R2 Test Score: 0.400417 Rho2 Test Score: -332884.0
0 2620 Train accuracy: 4.09031e-06 Test accuracy: 3.47007e-06 R2 Test Score: 0.410302 Rho2 Test Score: -339140.0
0 2630 Train accuracy: 3.41464e-06 Test accuracy: 3.59244e-06 R2 Test Score: 0.448254 Rho2 Test Score: -307003.0
0 2640 Train accuracy: 4.43466e-06 Test accuracy: 3.26351e-06 R2 Test Score: 0.454335 Rho2 Test Score: -311636.0
0 2650 Train accuracy: 3.71884e-06 Test accuracy: 3.39401e-06 R2 Test Score: 0.426047 Rho2 Test Score: -327952.0
0 2660 Train accuracy: 5.13293e-06 Test accuracy: 3.80153e-06 R2 Test Score: 0.413203 Rho2 Test Score: -319303.0
0 2670 Train accuracy: 3.99907e-06 Test accuracy: 4.16598e-06 R2 Test Score: 0.368332 Rho2 Test Score: -342290.0
0 2680 Train accuracy: 4.9817e-06 Test accuracy: 3.62396e-06 R2 Test Score: 0.428836 Rho2 Test Score: -315795.0
0 2690 Train accuracy: 5.71349e-06 Test accuracy: 3.48604e-06 R2 Test Score: 0.461489 Rho2 Test Score: -322597.0
0 2700 Train accuracy: 4.3579e-06 Test accuracy: 4.65179e-06 R2 Test Score: 0.375885 Rho2 Test Score: -342401.0
0 2710 Train accuracy: 3.90668e-06 Test accuracy: 3.82126e-06 R2 Test Score: 0.404535 Rho2 Test Score: -321203.0
0 2720 Train accuracy: 2.67891e-06 Test accuracy: 3.57722e-06 R2 Test Score: 0.430258 Rho2 Test Score: -314173.0
0 2730 Train accuracy: 3.65244e-06 Test accuracy: 4.21949e-06 R2 Test Score: 0.355452 Rho2 Test Score: -338531.0
0 2740 Train accuracy: 3.37404e-06 Test accuracy: 4.19194e-06 R2 Test Score: 0.303391 Rho2 Test Score: -366746.0
0 2750 Train accuracy: 4.21369e-06 Test accuracy: 4.28677e-06 R2 Test Score: 0.382239 Rho2 Test Score: -331115.0
0 2760 Train accuracy: 4.49593e-06 Test accuracy: 3.53412e-06 R2 Test Score: 0.435242 Rho2 Test Score: -324883.0
0 2770 Train accuracy: 6.70077e-06 Test accuracy: 4.09592e-06 R2 Test Score: 0.339364 Rho2 Test Score: -358211.0
0 2780 Train accuracy: 3.93417e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.371114 Rho2 Test Score: -353970.0
0 2790 Train accuracy: 4.39083e-06 Test accuracy: 3.46004e-06 R2 Test Score: 0.48565 Rho2 Test Score: -298184.0
0 2800 Train accuracy: 3.09425e-06 Test accuracy: 4.45549e-06 R2 Test Score: 0.374681 Rho2 Test Score: -327294.0
0 2810 Train accuracy: 3.07037e-06 Test accuracy: 3.89414e-06 R2 Test Score: 0.355383 Rho2 Test Score: -373214.0
0 2820 Train accuracy: 4.68334e-06 Test accuracy: 3.83695e-06 R2 Test Score: 0.337692 Rho2 Test Score: -375437.0
0 2830 Train accuracy: 4.36964e-06 Test accuracy: 4.10108e-06 R2 Test Score: 0.355712 Rho2 Test Score: -348139.0
0 2840 Train accuracy: 4.6401e-06 Test accuracy: 4.13223e-06 R2 Test Score: 0.393623 Rho2 Test Score: -343032.0
0 2850 Train accuracy: 4.69985e-06 Test accuracy: 3.49311e-06 R2 Test Score: 0.426135 Rho2 Test Score: -324264.0
0 2860 Train accuracy: 3.98055e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.402704 Rho2 Test Score: -328903.0
0 2870 Train accuracy: 4.09622e-06 Test accuracy: 3.93246e-06 R2 Test Score: 0.344632 Rho2 Test Score: -359561.0
0 2880 Train accuracy: 4.60259e-06 Test accuracy: 4.08907e-06 R2 Test Score: 0.340747 Rho2 Test Score: -371460.0
0 2890 Train accuracy: 4.08875e-06 Test accuracy: 3.33298e-06 R2 Test Score: 0.416559 Rho2 Test Score: -321026.0
0 2900 Train accuracy: 4.08866e-06 Test accuracy: 4.52972e-06 R2 Test Score: 0.26138 Rho2 Test Score: -418094.0
0 2910 Train accuracy: 5.11411e-06 Test accuracy: 3.90919e-06 R2 Test Score: 0.404385 Rho2 Test Score: -324589.0
0 2920 Train accuracy: 3.63953e-06 Test accuracy: 2.77057e-06 R2 Test Score: 0.505458 Rho2 Test Score: -292029.0
0 2930 Train accuracy: 3.93724e-06 Test accuracy: 3.69035e-06 R2 Test Score: 0.424497 Rho2 Test Score: -318974.0
0 2940 Train accuracy: 3.35738e-06 Test accuracy: 3.5466e-06 R2 Test Score: 0.398309 Rho2 Test Score: -344470.0
0 2950 Train accuracy: 4.73837e-06 Test accuracy: 4.97353e-06 R2 Test Score: 0.272813 Rho2 Test Score: -381539.0
0 2960 Train accuracy: 3.69229e-06 Test accuracy: 3.60166e-06 R2 Test Score: 0.413166 Rho2 Test Score: -328040.0
0 2970 Train accuracy: 3.28657e-06 Test accuracy: 3.64648e-06 R2 Test Score: 0.477415 Rho2 Test Score: -293516.0
0 2980 Train accuracy: 4.02328e-06 Test accuracy: 3.79328e-06 R2 Test Score: 0.388175 Rho2 Test Score: -343702.0
0 2990 Train accuracy: 4.25841e-06 Test accuracy: 3.98183e-06 R2 Test Score: 0.358599 Rho2 Test Score: -346479.0
0 3000 Train accuracy: 5.87594e-06 Test accuracy: 4.6302e-06 R2 Test Score: 0.232251 Rho2 Test Score: -430838.0
0 3010 Train accuracy: 2.67461e-06 Test accuracy: 3.56144e-06 R2 Test Score: 0.469755 Rho2 Test Score: -317065.0
0 3020 Train accuracy: 5.64083e-06 Test accuracy: 3.50286e-06 R2 Test Score: 0.413846 Rho2 Test Score: -316691.0
0 3030 Train accuracy: 4.03056e-06 Test accuracy: 4.21651e-06 R2 Test Score: 0.318362 Rho2 Test Score: -363365.0
0 3040 Train accuracy: 3.84514e-06 Test accuracy: 3.97359e-06 R2 Test Score: 0.39544 Rho2 Test Score: -322251.0
0 3050 Train accuracy: 3.14454e-06 Test accuracy: 3.22754e-06 R2 Test Score: 0.483777 Rho2 Test Score: -289913.0
0 3060 Train accuracy: 4.08245e-06 Test accuracy: 4.26933e-06 R2 Test Score: 0.379196 Rho2 Test Score: -339824.0
0 3070 Train accuracy: 4.78428e-06 Test accuracy: 4.65179e-06 R2 Test Score: 0.300511 Rho2 Test Score: -398206.0
0 3080 Train accuracy: 4.28221e-06 Test accuracy: 4.05785e-06 R2 Test Score: 0.356317 Rho2 Test Score: -354809.0
0 3090 Train accuracy: 3.34814e-06 Test accuracy: 3.18915e-06 R2 Test Score: 0.465481 Rho2 Test Score: -308882.0
0 3100 Train accuracy: 3.07684e-06 Test accuracy: 3.92386e-06 R2 Test Score: 0.349767 Rho2 Test Score: -364181.0
0 3110 Train accuracy: 2.83108e-06 Test accuracy: 3.48305e-06 R2 Test Score: 0.418856 Rho2 Test Score: -325813.0
0 3120 Train accuracy: 4.09128e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.35386 Rho2 Test Score: -352397.0
0 3130 Train accuracy: 6.43518e-06 Test accuracy: 3.9116e-06 R2 Test Score: 0.347627 Rho2 Test Score: -358301.0
0 3140 Train accuracy: 3.79814e-06 Test accuracy: 3.63275e-06 R2 Test Score: 0.420126 Rho2 Test Score: -320629.0
0 3150 Train accuracy: 4.64517e-06 Test accuracy: 3.73985e-06 R2 Test Score: 0.404904 Rho2 Test Score: -338724.0
0 3160 Train accuracy: 4.79367e-06 Test accuracy: 3.5391e-06 R2 Test Score: 0.397262 Rho2 Test Score: -325291.0
0 3170 Train accuracy: 3.48789e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.398929 Rho2 Test Score: -325768.0
0 3180 Train accuracy: 4.90559e-06 Test accuracy: 4.08718e-06 R2 Test Score: 0.379562 Rho2 Test Score: -334456.0
0 3190 Train accuracy: 6.00667e-06 Test accuracy: 4.34708e-06 R2 Test Score: 0.287787 Rho2 Test Score: -390218.0
0 3200 Train accuracy: 4.27442e-06 Test accuracy: 4.35195e-06 R2 Test Score: 0.325609 Rho2 Test Score: -373565.0
0 3210 Train accuracy: 4.76881e-06 Test accuracy: 3.88812e-06 R2 Test Score: 0.402164 Rho2 Test Score: -341132.0
0 3220 Train accuracy: 3.40328e-06 Test accuracy: 3.96328e-06 R2 Test Score: 0.359979 Rho2 Test Score: -343672.0
0 3230 Train accuracy: 4.85363e-06 Test accuracy: 3.20984e-06 R2 Test Score: 0.465115 Rho2 Test Score: -317015.0
0 3240 Train accuracy: 4.20231e-06 Test accuracy: 3.82046e-06 R2 Test Score: 0.41061 Rho2 Test Score: -328036.0
0 3250 Train accuracy: 4.67022e-06 Test accuracy: 3.72388e-06 R2 Test Score: 0.448559 Rho2 Test Score: -312990.0
0 3260 Train accuracy: 2.59698e-06 Test accuracy: 3.90016e-06 R2 Test Score: 0.388019 Rho2 Test Score: -342254.0
0 3270 Train accuracy: 4.50038e-06 Test accuracy: 4.03078e-06 R2 Test Score: 0.353444 Rho2 Test Score: -349311.0
0 3280 Train accuracy: 3.43651e-06 Test accuracy: 3.30412e-06 R2 Test Score: 0.51558 Rho2 Test Score: -285923.0
0 3290 Train accuracy: 6.71626e-06 Test accuracy: 4.74034e-06 R2 Test Score: 0.248607 Rho2 Test Score: -404711.0
0 3300 Train accuracy: 4.74504e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.345311 Rho2 Test Score: -352523.0
0 3310 Train accuracy: 3.29566e-06 Test accuracy: 3.56946e-06 R2 Test Score: 0.39841 Rho2 Test Score: -338043.0
0 3320 Train accuracy: 5.22791e-06 Test accuracy: 4.17395e-06 R2 Test Score: 0.37008 Rho2 Test Score: -350846.0
0 3330 Train accuracy: 5.46724e-06 Test accuracy: 4.08568e-06 R2 Test Score: 0.373656 Rho2 Test Score: -358316.0
0 3340 Train accuracy: 4.36398e-06 Test accuracy: 3.76022e-06 R2 Test Score: 0.412261 Rho2 Test Score: -323768.0
0 3350 Train accuracy: 3.91318e-06 Test accuracy: 3.93584e-06 R2 Test Score: 0.403619 Rho2 Test Score: -337936.0
0 3360 Train accuracy: 3.88401e-06 Test accuracy: 3.67568e-06 R2 Test Score: 0.404742 Rho2 Test Score: -329758.0
0 3370 Train accuracy: 4.6704e-06 Test accuracy: 3.32145e-06 R2 Test Score: 0.441814 Rho2 Test Score: -320083.0
0 3380 Train accuracy: 4.1184e-06 Test accuracy: 3.64095e-06 R2 Test Score: 0.416698 Rho2 Test Score: -333155.0
0 3390 Train accuracy: 5.26434e-06 Test accuracy: 4.11604e-06 R2 Test Score: 0.442529 Rho2 Test Score: -325629.0
0 3400 Train accuracy: 5.76661e-06 Test accuracy: 3.9337e-06 R2 Test Score: 0.38137 Rho2 Test Score: -347858.0
0 3410 Train accuracy: 4.18728e-06 Test accuracy: 4.42328e-06 R2 Test Score: 0.385733 Rho2 Test Score: -327362.0
0 3420 Train accuracy: 3.84525e-06 Test accuracy: 4.60269e-06 R2 Test Score: 0.315281 Rho2 Test Score: -385343.0
0 3430 Train accuracy: 4.63541e-06 Test accuracy: 4.32792e-06 R2 Test Score: 0.329673 Rho2 Test Score: -386286.0
0 3440 Train accuracy: 3.66007e-06 Test accuracy: 3.82158e-06 R2 Test Score: 0.395444 Rho2 Test Score: -351515.0
0 3450 Train accuracy: 3.38997e-06 Test accuracy: 3.93677e-06 R2 Test Score: 0.420987 Rho2 Test Score: -320719.0
0 3460 Train accuracy: 4.72577e-06 Test accuracy: 4.52584e-06 R2 Test Score: 0.391192 Rho2 Test Score: -356372.0
0 3470 Train accuracy: 3.78698e-06 Test accuracy: 3.96013e-06 R2 Test Score: 0.363016 Rho2 Test Score: -345995.0
0 3480 Train accuracy: 3.66447e-06 Test accuracy: 3.82146e-06 R2 Test Score: 0.348669 Rho2 Test Score: -355452.0
0 3490 Train accuracy: 3.62781e-06 Test accuracy: 3.3838e-06 R2 Test Score: 0.446627 Rho2 Test Score: -312298.0
0 3500 Train accuracy: 4.79871e-06 Test accuracy: 4.30083e-06 R2 Test Score: 0.258812 Rho2 Test Score: -398521.0
0 3510 Train accuracy: 6.2759e-06 Test accuracy: 4.2485e-06 R2 Test Score: 0.331935 Rho2 Test Score: -365512.0
0 3520 Train accuracy: 5.56891e-06 Test accuracy: 3.85001e-06 R2 Test Score: 0.415372 Rho2 Test Score: -322526.0
0 3530 Train accuracy: 5.68413e-06 Test accuracy: 4.52972e-06 R2 Test Score: 0.243199 Rho2 Test Score: -418165.0
0 3540 Train accuracy: 5.92573e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.323013 Rho2 Test Score: -355511.0
0 3550 Train accuracy: 4.48296e-06 Test accuracy: 4.07162e-06 R2 Test Score: 0.352903 Rho2 Test Score: -361454.0
0 3560 Train accuracy: 2.96694e-06 Test accuracy: 3.5535e-06 R2 Test Score: 0.406747 Rho2 Test Score: -331772.0
0 3570 Train accuracy: 5.87378e-06 Test accuracy: 3.51191e-06 R2 Test Score: 0.448442 Rho2 Test Score: -306147.0
0 3580 Train accuracy: 5.99562e-06 Test accuracy: 3.56512e-06 R2 Test Score: 0.424197 Rho2 Test Score: -335258.0
0 3590 Train accuracy: 5.10335e-06 Test accuracy: 3.38823e-06 R2 Test Score: 0.468762 Rho2 Test Score: -295888.0
0 3600 Train accuracy: 2.87402e-06 Test accuracy: 3.65721e-06 R2 Test Score: 0.438918 Rho2 Test Score: -313256.0
0 3610 Train accuracy: 4.07328e-06 Test accuracy: 4.38442e-06 R2 Test Score: 0.280947 Rho2 Test Score: -397792.0
0 3620 Train accuracy: 5.06682e-06 Test accuracy: 3.67345e-06 R2 Test Score: 0.379078 Rho2 Test Score: -344929.0
0 3630 Train accuracy: 4.25129e-06 Test accuracy: 3.38912e-06 R2 Test Score: 0.451159 Rho2 Test Score: -322802.0
0 3640 Train accuracy: 5.42152e-06 Test accuracy: 3.86063e-06 R2 Test Score: 0.403851 Rho2 Test Score: -318480.0
0 3650 Train accuracy: 3.78657e-06 Test accuracy: 3.99752e-06 R2 Test Score: 0.369196 Rho2 Test Score: -344684.0
0 3660 Train accuracy: 6.43743e-06 Test accuracy: 4.41914e-06 R2 Test Score: 0.345492 Rho2 Test Score: -363042.0
0 3670 Train accuracy: 3.45426e-06 Test accuracy: 4.25488e-06 R2 Test Score: 0.355001 Rho2 Test Score: -338292.0
0 3680 Train accuracy: 4.49071e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.255481 Rho2 Test Score: -412505.0
0 3690 Train accuracy: 4.08092e-06 Test accuracy: 5.74657e-06 R2 Test Score: 0.161406 Rho2 Test Score: -442067.0
0 3700 Train accuracy: 4.82393e-06 Test accuracy: 5.26214e-06 R2 Test Score: 0.216798 Rho2 Test Score: -442263.0
0 3710 Train accuracy: 5.96637e-06 Test accuracy: 3.79213e-06 R2 Test Score: 0.380488 Rho2 Test Score: -359900.0
0 3720 Train accuracy: 5.77867e-06 Test accuracy: 4.40595e-06 R2 Test Score: 0.319969 Rho2 Test Score: -373000.0
0 3730 Train accuracy: 5.86594e-06 Test accuracy: 4.73985e-06 R2 Test Score: 0.261321 Rho2 Test Score: -412355.0
0 3740 Train accuracy: 4.4729e-06 Test accuracy: 3.76908e-06 R2 Test Score: 0.403974 Rho2 Test Score: -328093.0
0 3750 Train accuracy: 3.7768e-06 Test accuracy: 3.89144e-06 R2 Test Score: 0.329977 Rho2 Test Score: -372000.0
0 3760 Train accuracy: 4.02628e-06 Test accuracy: 3.86854e-06 R2 Test Score: 0.354871 Rho2 Test Score: -353713.0
0 3770 Train accuracy: 3.54509e-06 Test accuracy: 3.89264e-06 R2 Test Score: 0.40252 Rho2 Test Score: -349904.0
0 3780 Train accuracy: 4.09325e-06 Test accuracy: 3.89553e-06 R2 Test Score: 0.448929 Rho2 Test Score: -320054.0
0 3790 Train accuracy: 4.49984e-06 Test accuracy: 4.1672e-06 R2 Test Score: 0.391659 Rho2 Test Score: -330642.0
0 3800 Train accuracy: 3.13564e-06 Test accuracy: 4.48359e-06 R2 Test Score: 0.341892 Rho2 Test Score: -359048.0
0 3810 Train accuracy: 5.22824e-06 Test accuracy: 4.22846e-06 R2 Test Score: 0.310501 Rho2 Test Score: -371177.0
0 3820 Train accuracy: 5.76787e-06 Test accuracy: 4.20725e-06 R2 Test Score: 0.311756 Rho2 Test Score: -372284.0
0 3830 Train accuracy: 7.17845e-06 Test accuracy: 5.26214e-06 R2 Test Score: 0.165357 Rho2 Test Score: -448123.0
0 3840 Train accuracy: 6.85725e-06 Test accuracy: 7.67551e-06 R2 Test Score: -0.292207 Rho2 Test Score: -738807.0
0 3850 Train accuracy: 4.41582e-06 Test accuracy: 4.50007e-06 R2 Test Score: 0.300833 Rho2 Test Score: -401021.0
0 3860 Train accuracy: 4.01847e-06 Test accuracy: 3.68786e-06 R2 Test Score: 0.479393 Rho2 Test Score: -298376.0
0 3870 Train accuracy: 4.30875e-06 Test accuracy: 3.78949e-06 R2 Test Score: 0.362208 Rho2 Test Score: -357285.0
0 3880 Train accuracy: 6.70556e-06 Test accuracy: 4.35923e-06 R2 Test Score: 0.299693 Rho2 Test Score: -392108.0
0 3890 Train accuracy: 4.59396e-06 Test accuracy: 4.05198e-06 R2 Test Score: 0.398192 Rho2 Test Score: -318526.0
0 3900 Train accuracy: 5.66701e-06 Test accuracy: 4.05568e-06 R2 Test Score: 0.366171 Rho2 Test Score: -347965.0
0 3910 Train accuracy: 3.50748e-06 Test accuracy: 3.68793e-06 R2 Test Score: 0.417282 Rho2 Test Score: -315025.0
0 3920 Train accuracy: 8.85206e-06 Test accuracy: 5.42053e-06 R2 Test Score: 0.0749843 Rho2 Test Score: -549572.0
0 3930 Train accuracy: 4.99614e-06 Test accuracy: 5.68835e-06 R2 Test Score: 0.188428 Rho2 Test Score: -444184.0
0 3940 Train accuracy: 4.26292e-06 Test accuracy: 5.2796e-06 R2 Test Score: 0.256035 Rho2 Test Score: -417401.0
0 3950 Train accuracy: 4.25484e-06 Test accuracy: 3.93191e-06 R2 Test Score: 0.395241 Rho2 Test Score: -322780.0
0 3960 Train accuracy: 4.75472e-06 Test accuracy: 4.19706e-06 R2 Test Score: 0.362619 Rho2 Test Score: -361946.0
0 3970 Train accuracy: 4.25064e-06 Test accuracy: 4.1776e-06 R2 Test Score: 0.414653 Rho2 Test Score: -325117.0
0 3980 Train accuracy: 4.0868e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.307863 Rho2 Test Score: -374233.0
0 3990 Train accuracy: 4.82593e-06 Test accuracy: 4.10763e-06 R2 Test Score: 0.299724 Rho2 Test Score: -393764.0
0 4000 Train accuracy: 6.72625e-06 Test accuracy: 4.05267e-06 R2 Test Score: 0.298058 Rho2 Test Score: -366178.0
0 4010 Train accuracy: 5.09598e-06 Test accuracy: 4.30946e-06 R2 Test Score: 0.306671 Rho2 Test Score: -369340.0
0 4020 Train accuracy: 6.47341e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.363998 Rho2 Test Score: -343182.0
0 4030 Train accuracy: 9.18754e-06 Test accuracy: 7.24652e-06 R2 Test Score: -0.271783 Rho2 Test Score: -711426.0
0 4040 Train accuracy: 7.07301e-06 Test accuracy: 4.70549e-06 R2 Test Score: 0.29316 Rho2 Test Score: -403369.0
0 4050 Train accuracy: 5.65175e-06 Test accuracy: 5.50406e-06 R2 Test Score: 0.11394 Rho2 Test Score: -484507.0
0 4060 Train accuracy: 4.15976e-06 Test accuracy: 3.43676e-06 R2 Test Score: 0.400591 Rho2 Test Score: -338168.0
0 4070 Train accuracy: 5.29287e-06 Test accuracy: 5.13131e-06 R2 Test Score: 0.244692 Rho2 Test Score: -440298.0
0 4080 Train accuracy: 7.99097e-06 Test accuracy: 4.31962e-06 R2 Test Score: 0.291959 Rho2 Test Score: -397195.0
0 4090 Train accuracy: 5.24003e-06 Test accuracy: 4.72693e-06 R2 Test Score: 0.225996 Rho2 Test Score: -423831.0
0 4100 Train accuracy: 3.19411e-06 Test accuracy: 3.8655e-06 R2 Test Score: 0.360882 Rho2 Test Score: -346864.0
0 4110 Train accuracy: 5.12082e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.366905 Rho2 Test Score: -361885.0
0 4120 Train accuracy: 4.45916e-06 Test accuracy: 3.91285e-06 R2 Test Score: 0.308898 Rho2 Test Score: -375876.0
0 4130 Train accuracy: 3.48978e-06 Test accuracy: 3.57135e-06 R2 Test Score: 0.443291 Rho2 Test Score: -316600.0
0 4140 Train accuracy: 6.2171e-06 Test accuracy: 5.25627e-06 R2 Test Score: 0.138168 Rho2 Test Score: -490025.0
0 4150 Train accuracy: 3.25766e-06 Test accuracy: 3.17687e-06 R2 Test Score: 0.375439 Rho2 Test Score: -350393.0
0 4160 Train accuracy: 5.35237e-06 Test accuracy: 4.40386e-06 R2 Test Score: 0.340589 Rho2 Test Score: -362461.0
0 4170 Train accuracy: 4.81715e-06 Test accuracy: 3.76673e-06 R2 Test Score: 0.353162 Rho2 Test Score: -353120.0
0 4180 Train accuracy: 5.06788e-06 Test accuracy: 3.91582e-06 R2 Test Score: 0.377499 Rho2 Test Score: -334292.0
0 4190 Train accuracy: 6.81345e-06 Test accuracy: 7.16381e-06 R2 Test Score: -0.0645244 Rho2 Test Score: -586651.0
0 4200 Train accuracy: 7.01007e-06 Test accuracy: 6.43677e-06 R2 Test Score: -0.0472223 Rho2 Test Score: -578755.0
0 4210 Train accuracy: 7.58448e-06 Test accuracy: 4.85307e-06 R2 Test Score: 0.219995 Rho2 Test Score: -433101.0
0 4220 Train accuracy: 4.61243e-06 Test accuracy: 3.67811e-06 R2 Test Score: 0.402326 Rho2 Test Score: -328156.0
0 4230 Train accuracy: 4.90418e-06 Test accuracy: 5.36513e-06 R2 Test Score: 0.162031 Rho2 Test Score: -473956.0
0 4240 Train accuracy: 8.57075e-06 Test accuracy: 4.59076e-06 R2 Test Score: 0.174694 Rho2 Test Score: -453642.0
0 4250 Train accuracy: 8.54116e-06 Test accuracy: 5.65238e-06 R2 Test Score: 0.153252 Rho2 Test Score: -471066.0
0 4260 Train accuracy: 5.58745e-06 Test accuracy: 5.64461e-06 R2 Test Score: 0.108187 Rho2 Test Score: -514093.0
0 4270 Train accuracy: 1.07643e-05 Test accuracy: 5.53072e-06 R2 Test Score: 0.173614 Rho2 Test Score: -485695.0
0 4280 Train accuracy: 5.57034e-06 Test accuracy: 4.81979e-06 R2 Test Score: 0.238137 Rho2 Test Score: -426299.0
0 4290 Train accuracy: 4.06685e-06 Test accuracy: 3.60271e-06 R2 Test Score: 0.381266 Rho2 Test Score: -356070.0
0 4300 Train accuracy: 7.68567e-06 Test accuracy: 4.75105e-06 R2 Test Score: 0.303853 Rho2 Test Score: -401447.0
0 4310 Train accuracy: 3.62102e-06 Test accuracy: 4.56327e-06 R2 Test Score: 0.317186 Rho2 Test Score: -361825.0
0 4320 Train accuracy: 5.683e-06 Test accuracy: 4.63009e-06 R2 Test Score: 0.257956 Rho2 Test Score: -412477.0
0 4330 Train accuracy: 6.00135e-06 Test accuracy: 4.63483e-06 R2 Test Score: 0.262773 Rho2 Test Score: -389141.0
0 4340 Train accuracy: 5.77531e-06 Test accuracy: 5.1913e-06 R2 Test Score: 0.210051 Rho2 Test Score: -436670.0
0 4350 Train accuracy: 7.63466e-06 Test accuracy: 6.62061e-06 R2 Test Score: -0.060323 Rho2 Test Score: -605618.0
0 4360 Train accuracy: 7.16346e-06 Test accuracy: 5.92176e-06 R2 Test Score: 0.054822 Rho2 Test Score: -542863.0
0 4370 Train accuracy: 6.37533e-06 Test accuracy: 4.35943e-06 R2 Test Score: 0.282815 Rho2 Test Score: -410480.0
0 4380 Train accuracy: 5.73653e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.354246 Rho2 Test Score: -354031.0
0 4390 Train accuracy: 3.41833e-06 Test accuracy: 3.30546e-06 R2 Test Score: 0.515329 Rho2 Test Score: -274126.0
0 4400 Train accuracy: 5.2028e-06 Test accuracy: 3.76618e-06 R2 Test Score: 0.41648 Rho2 Test Score: -320173.0
0 4410 Train accuracy: 6.05066e-06 Test accuracy: 4.76226e-06 R2 Test Score: 0.259012 Rho2 Test Score: -416967.0
0 4420 Train accuracy: 4.23918e-06 Test accuracy: 3.4634e-06 R2 Test Score: 0.463893 Rho2 Test Score: -308496.0
0 4430 Train accuracy: 4.3893e-06 Test accuracy: 4.5721e-06 R2 Test Score: 0.29642 Rho2 Test Score: -376515.0
0 4440 Train accuracy: 5.25387e-06 Test accuracy: 3.70046e-06 R2 Test Score: 0.429423 Rho2 Test Score: -313524.0
0 4450 Train accuracy: 4.60359e-06 Test accuracy: 4.14652e-06 R2 Test Score: 0.385928 Rho2 Test Score: -345664.0
0 4460 Train accuracy: 3.81368e-06 Test accuracy: 3.86711e-06 R2 Test Score: 0.360827 Rho2 Test Score: -362672.0
0 4470 Train accuracy: 4.86691e-06 Test accuracy: 4.88482e-06 R2 Test Score: 0.224635 Rho2 Test Score: -436562.0
0 4480 Train accuracy: 4.97577e-06 Test accuracy: 3.6493e-06 R2 Test Score: 0.392143 Rho2 Test Score: -351214.0
0 4490 Train accuracy: 3.8561e-06 Test accuracy: 4.31551e-06 R2 Test Score: 0.365689 Rho2 Test Score: -330535.0
0 4500 Train accuracy: 3.77214e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.340019 Rho2 Test Score: -371010.0
0 4510 Train accuracy: 4.08645e-06 Test accuracy: 4.22166e-06 R2 Test Score: 0.343791 Rho2 Test Score: -360852.0
0 4520 Train accuracy: 7.53955e-06 Test accuracy: 4.766e-06 R2 Test Score: 0.224303 Rho2 Test Score: -443090.0
0 4530 Train accuracy: 7.72607e-06 Test accuracy: 6.63329e-06 R2 Test Score: -0.01693 Rho2 Test Score: -560313.0
0 4540 Train accuracy: 1.00145e-05 Test accuracy: 8.60822e-06 R2 Test Score: -0.327228 Rho2 Test Score: -693465.0
0 4550 Train accuracy: 1.02796e-05 Test accuracy: 1.00918e-05 R2 Test Score: -0.647813 Rho2 Test Score: -868203.0
0 4560 Train accuracy: 1.10054e-05 Test accuracy: 1.05646e-05 R2 Test Score: -0.565177 Rho2 Test Score: -907144.0
0 4570 Train accuracy: 7.18482e-06 Test accuracy: 5.17962e-06 R2 Test Score: 0.196382 Rho2 Test Score: -462630.0
0 4580 Train accuracy: 5.60662e-06 Test accuracy: 4.73549e-06 R2 Test Score: 0.215506 Rho2 Test Score: -422420.0
0 4590 Train accuracy: 4.46947e-06 Test accuracy: 5.26468e-06 R2 Test Score: 0.247277 Rho2 Test Score: -433733.0
0 4600 Train accuracy: 5.61551e-06 Test accuracy: 4.59847e-06 R2 Test Score: 0.275229 Rho2 Test Score: -402044.0
0 4610 Train accuracy: 3.94336e-06 Test accuracy: 3.94461e-06 R2 Test Score: 0.358672 Rho2 Test Score: -344890.0
0 4620 Train accuracy: 4.2995e-06 Test accuracy: 4.46887e-06 R2 Test Score: 0.323484 Rho2 Test Score: -384137.0
0 4630 Train accuracy: 4.07534e-06 Test accuracy: 4.74751e-06 R2 Test Score: 0.337274 Rho2 Test Score: -356042.0
0 4640 Train accuracy: 5.65219e-06 Test accuracy: 5.14383e-06 R2 Test Score: 0.109117 Rho2 Test Score: -519809.0
0 4650 Train accuracy: 4.47779e-06 Test accuracy: 5.87049e-06 R2 Test Score: 0.175332 Rho2 Test Score: -481307.0
0 4660 Train accuracy: 6.02798e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.317126 Rho2 Test Score: -384386.0
0 4670 Train accuracy: 5.81226e-06 Test accuracy: 4.22039e-06 R2 Test Score: 0.322683 Rho2 Test Score: -359629.0
0 4680 Train accuracy: 3.88796e-06 Test accuracy: 4.76914e-06 R2 Test Score: 0.319501 Rho2 Test Score: -378808.0
0 4690 Train accuracy: 3.54034e-06 Test accuracy: 3.66071e-06 R2 Test Score: 0.433802 Rho2 Test Score: -306261.0
0 4700 Train accuracy: 6.35209e-06 Test accuracy: 5.38607e-06 R2 Test Score: 0.124508 Rho2 Test Score: -485289.0
0 4710 Train accuracy: 6.87652e-06 Test accuracy: 4.59204e-06 R2 Test Score: 0.339053 Rho2 Test Score: -366647.0
0 4720 Train accuracy: 2.62102e-06 Test accuracy: 4.06816e-06 R2 Test Score: 0.377848 Rho2 Test Score: -340570.0
0 4730 Train accuracy: 4.45768e-06 Test accuracy: 3.9815e-06 R2 Test Score: 0.390439 Rho2 Test Score: -345328.0
0 4740 Train accuracy: 5.01223e-06 Test accuracy: 4.53492e-06 R2 Test Score: 0.371373 Rho2 Test Score: -342072.0
0 4750 Train accuracy: 6.17984e-06 Test accuracy: 6.06787e-06 R2 Test Score: 0.00902963 Rho2 Test Score: -572009.0
0 4760 Train accuracy: 5.37362e-06 Test accuracy: 4.23956e-06 R2 Test Score: 0.305158 Rho2 Test Score: -388086.0
0 4770 Train accuracy: 6.22355e-06 Test accuracy: 5.38429e-06 R2 Test Score: 0.10126 Rho2 Test Score: -498472.0
0 4780 Train accuracy: 5.42385e-06 Test accuracy: 5.3686e-06 R2 Test Score: 0.163707 Rho2 Test Score: -473938.0
0 4790 Train accuracy: 5.43555e-06 Test accuracy: 4.31811e-06 R2 Test Score: 0.208325 Rho2 Test Score: -447925.0
0 4800 Train accuracy: 4.75614e-06 Test accuracy: 4.24853e-06 R2 Test Score: 0.335102 Rho2 Test Score: -375228.0
0 4810 Train accuracy: 4.51043e-06 Test accuracy: 4.47985e-06 R2 Test Score: 0.303328 Rho2 Test Score: -379496.0
0 4820 Train accuracy: 8.50208e-06 Test accuracy: 6.1389e-06 R2 Test Score: 0.0534968 Rho2 Test Score: -557200.0
0 4830 Train accuracy: 1.05594e-05 Test accuracy: 1.00628e-05 R2 Test Score: -0.559769 Rho2 Test Score: -834232.0
0 4840 Train accuracy: 1.11455e-05 Test accuracy: 7.6286e-06 R2 Test Score: -0.221129 Rho2 Test Score: -677062.0
0 4850 Train accuracy: 1.35976e-05 Test accuracy: 9.56332e-06 R2 Test Score: -0.545792 Rho2 Test Score: -858083.0
0 4860 Train accuracy: 1.13525e-05 Test accuracy: 7.81122e-06 R2 Test Score: -0.328705 Rho2 Test Score: -713888.0
0 4870 Train accuracy: 7.92201e-06 Test accuracy: 9.92928e-06 R2 Test Score: -0.6948 Rho2 Test Score: -951736.0
0 4880 Train accuracy: 1.04022e-05 Test accuracy: 9.39825e-06 R2 Test Score: -0.411965 Rho2 Test Score: -772204.0
0 4890 Train accuracy: 7.23391e-06 Test accuracy: 4.79244e-06 R2 Test Score: 0.291219 Rho2 Test Score: -411956.0
0 4900 Train accuracy: 6.09539e-06 Test accuracy: 5.80567e-06 R2 Test Score: 0.0981032 Rho2 Test Score: -486386.0
0 4910 Train accuracy: 6.72417e-06 Test accuracy: 5.43557e-06 R2 Test Score: 0.173282 Rho2 Test Score: -491211.0
0 4920 Train accuracy: 6.16302e-06 Test accuracy: 5.17011e-06 R2 Test Score: 0.181667 Rho2 Test Score: -467070.0
0 4930 Train accuracy: 7.25284e-06 Test accuracy: 4.48016e-06 R2 Test Score: 0.342284 Rho2 Test Score: -364650.0
0 4940 Train accuracy: 4.26706e-06 Test accuracy: 3.96909e-06 R2 Test Score: 0.386256 Rho2 Test Score: -336949.0
0 4950 Train accuracy: 3.53489e-06 Test accuracy: 4.30311e-06 R2 Test Score: 0.389005 Rho2 Test Score: -337355.0
0 4960 Train accuracy: 4.24221e-06 Test accuracy: 4.17476e-06 R2 Test Score: 0.393502 Rho2 Test Score: -336019.0
0 4970 Train accuracy: 4.69361e-06 Test accuracy: 4.41913e-06 R2 Test Score: 0.338117 Rho2 Test Score: -362457.0
0 4980 Train accuracy: 6.48916e-06 Test accuracy: 4.77853e-06 R2 Test Score: 0.236579 Rho2 Test Score: -431086.0
0 4990 Train accuracy: 4.57605e-06 Test accuracy: 4.40925e-06 R2 Test Score: 0.405272 Rho2 Test Score: -335320.0
0 5000 Train accuracy: 6.38861e-06 Test accuracy: 4.39059e-06 R2 Test Score: 0.267926 Rho2 Test Score: -397653.0
0 5010 Train accuracy: 7.13707e-06 Test accuracy: 5.21414e-06 R2 Test Score: 0.083218 Rho2 Test Score: -538659.0
0 5020 Train accuracy: 5.27068e-06 Test accuracy: 4.1817e-06 R2 Test Score: 0.43593 Rho2 Test Score: -306965.0
0 5030 Train accuracy: 5.02256e-06 Test accuracy: 3.6991e-06 R2 Test Score: 0.433625 Rho2 Test Score: -326313.0
0 5040 Train accuracy: 4.56016e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.30806 Rho2 Test Score: -390585.0
0 5050 Train accuracy: 7.01695e-06 Test accuracy: 5.50438e-06 R2 Test Score: 0.213716 Rho2 Test Score: -437536.0
0 5060 Train accuracy: 3.26476e-06 Test accuracy: 3.62615e-06 R2 Test Score: 0.427827 Rho2 Test Score: -315091.0
0 5070 Train accuracy: 3.64979e-06 Test accuracy: 3.60105e-06 R2 Test Score: 0.434233 Rho2 Test Score: -311311.0
0 5080 Train accuracy: 5.52569e-06 Test accuracy: 3.83746e-06 R2 Test Score: 0.359146 Rho2 Test Score: -365254.0
0 5090 Train accuracy: 4.56835e-06 Test accuracy: 3.95908e-06 R2 Test Score: 0.351077 Rho2 Test Score: -346638.0
0 5100 Train accuracy: 5.97568e-06 Test accuracy: 4.37601e-06 R2 Test Score: 0.240514 Rho2 Test Score: -417091.0
0 5110 Train accuracy: 5.26517e-06 Test accuracy: 5.16816e-06 R2 Test Score: 0.238259 Rho2 Test Score: -422115.0
0 5120 Train accuracy: 5.44693e-06 Test accuracy: 4.65179e-06 R2 Test Score: 0.226463 Rho2 Test Score: -444881.0
0 5130 Train accuracy: 7.21475e-06 Test accuracy: 6.1969e-06 R2 Test Score: -0.020346 Rho2 Test Score: -593405.0
0 5140 Train accuracy: 5.01319e-06 Test accuracy: 4.95696e-06 R2 Test Score: 0.294073 Rho2 Test Score: -401668.0
0 5150 Train accuracy: 5.98054e-06 Test accuracy: 7.45712e-06 R2 Test Score: -0.275613 Rho2 Test Score: -715586.0
0 5160 Train accuracy: 4.73151e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.352359 Rho2 Test Score: -354130.0
0 5170 Train accuracy: 4.18631e-06 Test accuracy: 3.82471e-06 R2 Test Score: 0.366075 Rho2 Test Score: -354643.0
0 5180 Train accuracy: 5.90285e-06 Test accuracy: 3.9359e-06 R2 Test Score: 0.39465 Rho2 Test Score: -331704.0
0 5190 Train accuracy: 3.75263e-06 Test accuracy: 4.88843e-06 R2 Test Score: 0.191091 Rho2 Test Score: -465067.0
0 5200 Train accuracy: 6.08986e-06 Test accuracy: 6.28045e-06 R2 Test Score: 0.0733103 Rho2 Test Score: -549090.0
0 5210 Train accuracy: 7.64817e-06 Test accuracy: 5.56841e-06 R2 Test Score: 0.201634 Rho2 Test Score: -440958.0
0 5220 Train accuracy: 8.01348e-06 Test accuracy: 6.06183e-06 R2 Test Score: -0.00562859 Rho2 Test Score: -543347.0
0 5230 Train accuracy: 7.66626e-06 Test accuracy: 4.19475e-06 R2 Test Score: 0.328635 Rho2 Test Score: -383525.0
0 5240 Train accuracy: 7.51342e-06 Test accuracy: 5.26898e-06 R2 Test Score: 0.130992 Rho2 Test Score: -507858.0
0 5250 Train accuracy: 4.67254e-06 Test accuracy: 3.27341e-06 R2 Test Score: 0.432386 Rho2 Test Score: -320710.0
0 5260 Train accuracy: 4.31227e-06 Test accuracy: 3.49669e-06 R2 Test Score: 0.414621 Rho2 Test Score: -345369.0
0 5270 Train accuracy: 4.23734e-06 Test accuracy: 3.60433e-06 R2 Test Score: 0.395061 Rho2 Test Score: -339804.0
0 5280 Train accuracy: 5.99166e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.305246 Rho2 Test Score: -365362.0
0 5290 Train accuracy: 3.59639e-06 Test accuracy: 3.24429e-06 R2 Test Score: 0.484319 Rho2 Test Score: -303417.0
0 5300 Train accuracy: 4.3592e-06 Test accuracy: 3.30663e-06 R2 Test Score: 0.443809 Rho2 Test Score: -317700.0
0 5310 Train accuracy: 7.44655e-06 Test accuracy: 4.90139e-06 R2 Test Score: 0.215313 Rho2 Test Score: -439596.0
0 5320 Train accuracy: 3.86122e-06 Test accuracy: 3.18429e-06 R2 Test Score: 0.511816 Rho2 Test Score: -271741.0
0 5330 Train accuracy: 4.28757e-06 Test accuracy: 3.45598e-06 R2 Test Score: 0.444685 Rho2 Test Score: -310978.0
0 5340 Train accuracy: 7.95649e-06 Test accuracy: 4.95427e-06 R2 Test Score: 0.224482 Rho2 Test Score: -414531.0
0 5350 Train accuracy: 7.2961e-06 Test accuracy: 8.24565e-06 R2 Test Score: -0.199044 Rho2 Test Score: -646333.0
0 5360 Train accuracy: 4.58943e-06 Test accuracy: 3.7425e-06 R2 Test Score: 0.330432 Rho2 Test Score: -376149.0
0 5370 Train accuracy: 4.82116e-06 Test accuracy: 5.68939e-06 R2 Test Score: 0.119719 Rho2 Test Score: -509944.0
0 5380 Train accuracy: 4.79501e-06 Test accuracy: 3.63764e-06 R2 Test Score: 0.421217 Rho2 Test Score: -315732.0
0 5390 Train accuracy: 5.05094e-06 Test accuracy: 4.22448e-06 R2 Test Score: 0.401237 Rho2 Test Score: -328715.0
0 5400 Train accuracy: 6.73951e-06 Test accuracy: 5.56024e-06 R2 Test Score: 0.110887 Rho2 Test Score: -501106.0
0 5410 Train accuracy: 3.82257e-06 Test accuracy: 5.34581e-06 R2 Test Score: 0.201302 Rho2 Test Score: -440850.0
0 5420 Train accuracy: 4.83978e-06 Test accuracy: 4.11758e-06 R2 Test Score: 0.235113 Rho2 Test Score: -404850.0
0 5430 Train accuracy: 5.9433e-06 Test accuracy: 4.64823e-06 R2 Test Score: 0.267645 Rho2 Test Score: -423777.0
0 5440 Train accuracy: 7.60139e-06 Test accuracy: 6.32403e-06 R2 Test Score: 0.0158069 Rho2 Test Score: -570103.0
0 5450 Train accuracy: 6.86729e-06 Test accuracy: 5.33492e-06 R2 Test Score: 0.163354 Rho2 Test Score: -475597.0
0 5460 Train accuracy: 1.12404e-05 Test accuracy: 8.86075e-06 R2 Test Score: -0.362397 Rho2 Test Score: -756297.0
0 5470 Train accuracy: 6.14128e-06 Test accuracy: 4.46869e-06 R2 Test Score: 0.370137 Rho2 Test Score: -351218.0
0 5480 Train accuracy: 4.05176e-06 Test accuracy: 4.43065e-06 R2 Test Score: 0.283751 Rho2 Test Score: -418081.0
0 5490 Train accuracy: 5.51896e-06 Test accuracy: 6.1633e-06 R2 Test Score: 0.10962 Rho2 Test Score: -474626.0
0 5500 Train accuracy: 4.47402e-06 Test accuracy: 4.7743e-06 R2 Test Score: 0.293694 Rho2 Test Score: -394117.0
0 5510 Train accuracy: 5.38167e-06 Test accuracy: 4.74622e-06 R2 Test Score: 0.152818 Rho2 Test Score: -462951.0
0 5520 Train accuracy: 7.00664e-06 Test accuracy: 4.72508e-06 R2 Test Score: 0.246876 Rho2 Test Score: -432144.0
0 5530 Train accuracy: 9.64199e-06 Test accuracy: 1.0471e-05 R2 Test Score: -0.679872 Rho2 Test Score: -929020.0
0 5540 Train accuracy: 1.39834e-05 Test accuracy: 1.0536e-05 R2 Test Score: -0.744589 Rho2 Test Score: -1.00919e+06
0 5550 Train accuracy: 9.7037e-06 Test accuracy: 6.70515e-06 R2 Test Score: -0.0847446 Rho2 Test Score: -609942.0
0 5560 Train accuracy: 8.39063e-06 Test accuracy: 8.5268e-06 R2 Test Score: -0.468165 Rho2 Test Score: -811906.0
0 5570 Train accuracy: 6.30832e-06 Test accuracy: 5.785e-06 R2 Test Score: 0.0894905 Rho2 Test Score: -522637.0
0 5580 Train accuracy: 6.11626e-06 Test accuracy: 7.05321e-06 R2 Test Score: -0.0966561 Rho2 Test Score: -599513.0
0 5590 Train accuracy: 9.89845e-06 Test accuracy: 6.58077e-06 R2 Test Score: 0.042736 Rho2 Test Score: -540211.0
0 5600 Train accuracy: 6.4756e-06 Test accuracy: 8.58173e-06 R2 Test Score: -0.452134 Rho2 Test Score: -806292.0
0 5610 Train accuracy: 4.41343e-06 Test accuracy: 4.34662e-06 R2 Test Score: 0.261599 Rho2 Test Score: -411553.0
0 5620 Train accuracy: 6.20841e-06 Test accuracy: 5.34263e-06 R2 Test Score: 0.205103 Rho2 Test Score: -441320.0
0 5630 Train accuracy: 4.58817e-06 Test accuracy: 3.33981e-06 R2 Test Score: 0.430783 Rho2 Test Score: -321335.0
0 5640 Train accuracy: 5.49852e-06 Test accuracy: 5.26705e-06 R2 Test Score: 0.1749 Rho2 Test Score: -477304.0
0 5650 Train accuracy: 8.54209e-06 Test accuracy: 7.86487e-06 R2 Test Score: -0.271675 Rho2 Test Score: -690057.0
0 5660 Train accuracy: 5.49789e-06 Test accuracy: 4.75932e-06 R2 Test Score: 0.282645 Rho2 Test Score: -382756.0
0 5670 Train accuracy: 6.19575e-06 Test accuracy: 5.02764e-06 R2 Test Score: 0.188185 Rho2 Test Score: -456417.0
0 5680 Train accuracy: 3.71724e-06 Test accuracy: 3.20566e-06 R2 Test Score: 0.511807 Rho2 Test Score: -269083.0
0 5690 Train accuracy: 7.32972e-06 Test accuracy: 6.58522e-06 R2 Test Score: 0.0214506 Rho2 Test Score: -528910.0
0 5700 Train accuracy: 4.12154e-06 Test accuracy: 4.38586e-06 R2 Test Score: 0.382762 Rho2 Test Score: -346155.0
0 5710 Train accuracy: 6.46051e-06 Test accuracy: 4.77388e-06 R2 Test Score: 0.199485 Rho2 Test Score: -455184.0
0 5720 Train accuracy: 3.08656e-06 Test accuracy: 3.53343e-06 R2 Test Score: 0.465807 Rho2 Test Score: -308446.0
0 5730 Train accuracy: 7.05543e-06 Test accuracy: 5.75192e-06 R2 Test Score: 0.147568 Rho2 Test Score: -477905.0
0 5740 Train accuracy: 4.37807e-06 Test accuracy: 4.08033e-06 R2 Test Score: 0.392976 Rho2 Test Score: -332393.0
0 5750 Train accuracy: 4.73482e-06 Test accuracy: 4.40765e-06 R2 Test Score: 0.268976 Rho2 Test Score: -410652.0
0 5760 Train accuracy: 5.69303e-06 Test accuracy: 5.29544e-06 R2 Test Score: 0.175511 Rho2 Test Score: -476642.0
0 5770 Train accuracy: 4.3092e-06 Test accuracy: 5.66999e-06 R2 Test Score: 0.103739 Rho2 Test Score: -490391.0
0 5780 Train accuracy: 4.65119e-06 Test accuracy: 4.88919e-06 R2 Test Score: 0.251596 Rho2 Test Score: -421272.0
0 5790 Train accuracy: 5.4037e-06 Test accuracy: 4.46869e-06 R2 Test Score: 0.344575 Rho2 Test Score: -355292.0
0 5800 Train accuracy: 8.74832e-06 Test accuracy: 6.52869e-06 R2 Test Score: -0.161157 Rho2 Test Score: -658612.0
0 5810 Train accuracy: 7.09098e-06 Test accuracy: 6.81542e-06 R2 Test Score: -0.0320765 Rho2 Test Score: -585926.0
0 5820 Train accuracy: 7.37513e-06 Test accuracy: 6.50021e-06 R2 Test Score: 0.0235062 Rho2 Test Score: -528129.0
0 5830 Train accuracy: 5.36484e-06 Test accuracy: 6.07715e-06 R2 Test Score: 0.0763629 Rho2 Test Score: -499277.0

Training and Evaluating Our Model

add progress bar through python logging



In [ ]:

    
import logging
logging.getLogger().setLevel(logging.INFO)



In [ ]:

    
# Training Our Model
nFitSteps = 100000
start = time()
wrap  = regressor.fit(input_fn=train_input_fn, steps=nFitSteps)
print('TF Regressor took {} seconds'.format(time()-start))



In [ ]:

    
# Evaluating Our Model
print('Evaluating ...')
results = regressor.evaluate(input_fn=test_input_fn, steps=1)

for key in sorted(results):
    print("{}: {}".format(key, results[key]))

print("Val Acc: {:.3f}".format((1-results['loss'])*100))

Track Scalable Growth

Shrunk data set to 23559 Training samples and 7853 Val/Test samples

n_iters	time (s)	val acc	multicore	gpu
100	5.869	6.332	yes	no
200	6.380	13.178	yes	no
500	8.656	54.220	yes	no
1000	12.170	66.596	yes	no
2000	19.891	62.996	yes	no
5000	43.589	76.586	yes	no
10000	80.581	66.872	yes	no
20000	162.435	78.927	yes	no
50000	535.584	75.493	yes	no
100000	1062.656	73.162	yes	no



In [ ]:

    
nItersList = [100,200,500,1000,2000,5000,10000,20000,50000,100000]
rtimesList = [5.869, 6.380, 8.656, 12.170, 19.891, 43.589, 80.581, 162.435, 535.584, 1062.656]
valAccList = [6.332, 13.178, 54.220, 66.596, 62.996, 76.586, 66.872, 78.927, 75.493, 73.162]



In [ ]:

    
plt.loglog(nItersList, rtimesList,'o-');
plt.twinx()
plt.semilogx(nItersList, valAccList,'o-', color='orange');

Predicting output for test data

Most of the time prediction script would be separate from training script (we need not to train on same data again) but I am providing both in same script here; as I am not sure if we can create multiple notebook and somehow share data between them in Kaggle.



In [ ]:

    
def de_median(x):
    return x - np.median(x)



In [ ]:

    
predicted_output = list(regressor.predict(input_fn=test_input_fn))
# x = list(predicted_output)

# print([predicted_output() for _ in range(10)]) plt.plot((predicted_output - np.median(predicted_output)) / np.std(predicted_output),'.',alpha=0.1); plt.plot((test_df['flux'].values - np.median(test_df['flux'].values)) / np.std(test_df['flux'].values),'.',alpha=0.1);

plt.plot(de_median(x - test_df['flux'].values)/x,'.',alpha=0.1); plt.ylim(-1.,1.);



In [ ]:

    
r2_score(test_df['flux'].values,predicted_output)*100



In [ ]:

    
print('Full notebook took {} seconds'.format(time()-start0))

saveDir = 'tfSaveModels' regressor.export_savedmodel(saveDir, regressor)

saveDir = 'tfSaveModels' reg_args = {'feature_columns': fc, 'hidden_units': hu_array, ...} regressor = tf.contrib.learn.DNNRegressor(**reg_args) pickle.dump(reg_args, open('reg_args.pkl', 'wb'))

reg_args = pickle.load(open('reg_args.pkl', 'rb')) # On another machine and so my model dir path changed: reg_args['model_dir'] = NEW_MODEL_DIR regressor = tf.contrib.learn.DNNRegressor(**reg_args)

	pix1	pix2	pix3	pix4	pix5	pix6	pix7	pix8	pix9
0	577.447021	3465.876709	1118.598145	550.165466	2460.376953	994.374207	141.741592	521.385254	694.330688
1	569.863098	3387.739258	1087.530762	556.717407	2552.070557	1021.892700	134.061081	511.347778	666.346069
2	552.641235	3405.411377	1082.131104	558.981445	2560.040771	1058.485352	146.488220	513.809570	691.019653
3	571.821167	3340.533691	1073.962036	568.324768	2643.155273	1024.431641	147.687546	525.451538	727.683472
4	538.292114	3248.569336	1021.301208	548.598145	2691.563965	1066.199707	154.170990	541.407532	718.537537
5	553.332214	3183.050293	1026.863281	578.003784	2679.043457	1085.607422	154.789886	517.848328	722.456909
6	541.202332	3137.938232	1035.948364	589.566528	2743.923584	1072.393555	158.401291	539.485718	724.184937
7	547.699829	3057.428467	1034.417603	590.498108	2818.238770	1118.098633	167.132004	540.160950	760.767334
8	534.724976	3015.335205	1007.406494	574.792358	2876.417236	1141.010498	158.403992	527.611633	765.304077
9	529.423950	3010.478027	973.444946	578.736755	2875.608887	1106.352905	163.936325	540.751587	746.721191
10	543.917847	2959.855225	1002.942017	600.163574	2921.819580	1124.441772	153.599976	548.917603	735.936890
11	415.543091	1085.521729	446.313171	1641.077881	4778.175781	319.484497	418.018860	810.050232	567.318176
12	386.626038	990.775635	455.084015	1653.742676	4831.918457	318.239410	428.828278	836.078308	576.660156
13	352.204071	909.555847	448.546478	1697.422363	4942.354004	315.904175	434.170258	822.891968	576.275024
14	356.336823	830.570557	446.085022	1725.667725	4915.351074	312.517944	455.358154	826.022034	565.244507
15	346.803467	827.159546	445.986084	1783.878418	4926.551270	299.384399	454.638336	856.649109	572.549500
16	319.600525	778.882324	452.693604	1766.618774	4900.985352	295.125763	455.190002	873.377808	581.590820
17	318.834503	772.429138	450.526428	1804.441284	4888.824219	292.672455	460.712219	873.199463	557.806458
18	304.534882	747.088501	433.972229	1812.826172	4949.355469	292.384491	459.964996	883.855896	556.133789
19	297.334167	711.377441	427.756256	1824.887939	4977.011230	288.222961	442.602753	883.110352	578.062012
20	291.785248	696.303345	434.420410	1808.774414	4967.067383	288.906006	465.871857	848.896790	584.660889
21	275.729004	684.407776	436.270630	1839.743286	4958.617188	292.503906	473.751465	874.837891	576.924805
22	267.109802	668.616150	428.174561	1800.920166	5008.872559	289.184784	474.868591	872.165161	579.913147
23	272.932220	664.508301	437.244934	1776.687744	5002.353027	288.002045	470.427643	882.427673	600.059509
24	269.932495	641.567139	432.846771	1789.313354	5064.815918	303.351807	484.210266	894.807068	602.440247
25	265.979431	648.039856	432.684937	1755.147705	5088.369141	288.072937	482.552338	882.221191	592.758179
26	564.362610	1541.839600	431.914764	1446.160400	4362.256348	345.700531	401.668152	821.724915	520.839478
27	506.248077	1415.009277	443.126648	1487.735840	4414.598633	347.017731	393.479889	795.969360	549.537842
28	515.429810	1315.711914	448.500977	1547.112671	4509.289062	323.252441	394.831543	821.549072	529.604004
29	476.979919	1244.519043	429.070221	1599.996094	4529.244629	330.517914	398.165985	820.598389	528.216064
...	...	...	...	...	...	...	...	...	...
785255	475.147705	2842.766357	591.782959	633.689026	3837.167969	771.729736	238.382645	637.508911	712.187988
785256	475.464050	2835.344727	583.647827	640.641846	3873.379639	750.115234	248.780396	651.230103	716.854736
785257	481.847534	2817.340820	578.740479	637.754578	3877.218262	781.961548	241.087128	650.161316	733.283325
785258	465.787201	2804.904053	569.343262	647.821289	3857.104492	767.500122	240.323898	628.991760	726.848083
785259	473.557251	2804.657471	561.592529	633.638123	3901.351562	773.559692	231.294861	641.328003	733.939087
785260	476.991699	2791.731445	579.058228	643.239990	3923.991455	765.078247	235.408035	661.097107	718.653198
785261	476.635651	2750.070068	563.718689	635.713318	3949.495361	772.300659	246.790222	632.139221	730.945618
785262	481.483948	2736.324219	567.592407	660.586548	3944.277100	766.693787	241.954529	633.959961	734.623413
785263	472.956024	2736.230225	562.863098	659.623413	4004.276367	764.286011	239.283524	660.717346	713.998047
785264	466.867065	2732.458008	569.005798	666.330994	4024.462891	769.562134	245.346207	663.933350	725.252869
785265	480.046295	2715.175537	557.256165	645.355225	4037.946533	742.447998	258.840576	654.178589	706.549988
785266	475.449463	2769.531494	536.281189	655.740417	4031.584229	730.941650	255.209244	658.127136	719.966553
785267	502.837372	2716.243408	554.105591	677.264648	3980.764404	730.059265	259.389648	667.066284	708.509277
785268	480.399200	2769.769531	562.770813	679.593140	3983.617920	712.878052	243.168762	654.615112	689.989075
785269	480.807373	2759.118408	561.283997	699.134888	4009.483398	721.416626	262.486328	654.512329	713.022461
785270	523.245300	2958.067139	543.905029	697.710938	3776.409424	666.272888	263.427338	651.643799	670.494263
785271	495.875732	2955.598145	552.345215	657.297363	3747.275879	703.948242	250.976898	683.805420	657.604126
785272	516.237366	2964.318604	577.262634	668.211304	3793.730225	709.129517	256.492645	667.471802	683.404236
785273	501.681335	2919.767578	550.606995	638.708374	3802.873535	717.027649	250.493073	644.383301	702.473267
785274	510.000214	2995.055420	551.373047	660.349304	3745.079590	722.202332	247.733948	642.584961	695.485779
785275	521.931641	3013.854004	554.299988	631.953613	3710.329346	728.134094	243.888962	643.002563	691.181763
785276	495.807709	2989.654053	585.109985	639.402100	3714.461670	749.341003	235.572479	646.545593	676.286865
785277	512.751160	2958.394531	582.242310	652.946533	3707.906250	729.102295	234.076538	652.182556	688.491699
785278	504.160034	2989.499268	587.763611	658.194824	3695.002930	736.696594	231.465927	651.476196	681.635620
785279	516.517090	2976.608643	586.910278	637.642395	3714.598877	752.355835	235.402084	637.375977	700.220703
785280	506.839264	3055.761230	596.532349	647.959351	3610.522217	731.254211	234.202530	658.012695	675.393433
785281	495.875641	3055.138428	590.330444	623.768982	3686.792236	732.337524	234.055603	654.529114	690.857361
785282	510.378479	3120.925781	585.978516	623.555176	3621.902100	728.794128	230.795486	644.155518	694.674377
785283	520.521362	3118.752197	582.462891	632.206604	3600.458984	756.452515	224.326355	641.459534	696.739502
785284	520.870972	3144.105225	609.523743	617.985229	3561.716309	752.712402	225.764816	620.627991	680.224121

	pix1	pix2	pix3	pix4	pix5	pix6	pix7	pix8	pix9
0	0.054868	0.329321	0.106287	0.052276	0.233781	0.094484	0.013468	0.049541	0.065974
1	0.054337	0.323024	0.103697	0.053084	0.243342	0.097438	0.012783	0.048758	0.063537
2	0.052289	0.322207	0.102387	0.052889	0.242221	0.100150	0.013860	0.048615	0.065382
3	0.053828	0.314461	0.101097	0.053499	0.248813	0.096435	0.013903	0.049463	0.068500
4	0.051126	0.308546	0.097002	0.052105	0.255642	0.101267	0.014643	0.051422	0.068246
5	0.052693	0.303119	0.097787	0.055043	0.255123	0.103381	0.014740	0.049314	0.068799
6	0.051333	0.297631	0.098259	0.055920	0.260259	0.101716	0.015024	0.051170	0.068688
7	0.051502	0.287502	0.097271	0.055527	0.265011	0.105139	0.015716	0.050794	0.071538
8	0.050441	0.284439	0.095029	0.054221	0.271334	0.107632	0.014942	0.049770	0.072192
9	0.050299	0.286019	0.092485	0.054984	0.273205	0.105112	0.015575	0.051376	0.070944
10	0.051354	0.279453	0.094692	0.056664	0.275862	0.106164	0.014502	0.051826	0.069483
11	0.039645	0.103565	0.042581	0.156569	0.455867	0.030481	0.039882	0.077284	0.054126
12	0.036899	0.094558	0.043433	0.157831	0.461151	0.030372	0.040927	0.079794	0.055036
13	0.033545	0.086630	0.042721	0.161670	0.470731	0.030088	0.041352	0.078376	0.054887
14	0.034154	0.079609	0.042756	0.165402	0.471128	0.029954	0.043645	0.079173	0.054178
15	0.032986	0.078675	0.042420	0.169673	0.468588	0.028476	0.043243	0.081480	0.054458
16	0.030660	0.074720	0.043428	0.169475	0.470161	0.028312	0.043667	0.083785	0.055793
17	0.030600	0.074133	0.043239	0.173180	0.469202	0.028089	0.044217	0.083805	0.053535
18	0.029170	0.071559	0.041568	0.173640	0.474071	0.028006	0.044057	0.084660	0.053269
19	0.028507	0.068203	0.041011	0.174959	0.477166	0.027633	0.042434	0.084667	0.055421
20	0.028092	0.067038	0.041825	0.174144	0.478215	0.027815	0.044853	0.081729	0.056289
21	0.026480	0.065728	0.041898	0.176681	0.476205	0.028091	0.045497	0.084016	0.055405
22	0.025709	0.064353	0.041211	0.173335	0.482094	0.027833	0.045705	0.083944	0.055815
23	0.026257	0.063928	0.042064	0.170923	0.481243	0.027707	0.045257	0.084893	0.057728
24	0.025749	0.061199	0.041289	0.170683	0.483133	0.028937	0.046189	0.085356	0.057467
25	0.025487	0.062098	0.041461	0.168185	0.487587	0.027604	0.046240	0.084538	0.056800
26	0.054076	0.147736	0.041385	0.138568	0.417982	0.033124	0.038487	0.078736	0.049906
27	0.048900	0.136680	0.042803	0.143705	0.426419	0.033519	0.038007	0.076885	0.053081
28	0.049535	0.126447	0.043103	0.148685	0.433365	0.031066	0.037945	0.078955	0.050898
29	0.046052	0.120159	0.041427	0.154480	0.437299	0.031912	0.038443	0.079229	0.050999
...	...	...	...	...	...	...	...	...	...
785255	0.044239	0.264681	0.055099	0.059001	0.357266	0.071853	0.022195	0.059356	0.066309
785256	0.044125	0.263130	0.054165	0.059454	0.359463	0.069613	0.023088	0.060436	0.066527
785257	0.044618	0.260880	0.053590	0.059055	0.359022	0.072408	0.022324	0.060203	0.067900
785258	0.043496	0.261929	0.053167	0.060495	0.360187	0.071671	0.022442	0.058737	0.067875
785259	0.044032	0.260779	0.052217	0.058916	0.362750	0.071926	0.021506	0.059631	0.068242
785260	0.044185	0.258607	0.053640	0.059585	0.363492	0.070872	0.021807	0.061240	0.066571
785261	0.044306	0.255635	0.052401	0.059093	0.367128	0.071790	0.022941	0.058761	0.067946
785262	0.044716	0.254128	0.052714	0.061350	0.366313	0.071204	0.022471	0.058877	0.068226
785263	0.043735	0.253021	0.052048	0.060996	0.370278	0.070674	0.022127	0.061097	0.066024
785264	0.042977	0.251533	0.052379	0.061338	0.370467	0.070841	0.022585	0.061118	0.066762
785265	0.044458	0.251456	0.051608	0.059767	0.373960	0.068759	0.023972	0.060584	0.065435
785266	0.043890	0.255661	0.049505	0.060533	0.372163	0.067475	0.023559	0.060753	0.066462
785267	0.046575	0.251592	0.051324	0.062732	0.368718	0.067622	0.024026	0.061787	0.065626
785268	0.044577	0.257012	0.052221	0.063061	0.369648	0.066149	0.022564	0.060743	0.064025
785269	0.044268	0.254033	0.051678	0.064370	0.369154	0.066421	0.024167	0.060261	0.065648
785270	0.048669	0.275139	0.050590	0.064896	0.351255	0.061972	0.024502	0.060611	0.062365
785271	0.046323	0.276102	0.051598	0.061403	0.350058	0.065761	0.023445	0.063879	0.061431
785272	0.047640	0.273556	0.053271	0.061664	0.350096	0.065440	0.023670	0.061596	0.063066
785273	0.046764	0.272163	0.051324	0.059536	0.354481	0.066837	0.023349	0.060065	0.065480
785274	0.047354	0.278096	0.051196	0.061315	0.347737	0.067058	0.023003	0.059665	0.064577
785275	0.048603	0.280657	0.051618	0.058849	0.345514	0.067805	0.022711	0.059878	0.064364
785276	0.046198	0.278569	0.054519	0.059578	0.346105	0.069822	0.021950	0.060244	0.063015
785277	0.047840	0.276019	0.054323	0.060920	0.345948	0.068025	0.021839	0.060849	0.064236
785278	0.046960	0.278458	0.054748	0.061308	0.344173	0.068620	0.021560	0.060682	0.063491
785279	0.048014	0.276697	0.054558	0.059273	0.345299	0.069937	0.021882	0.059249	0.065091
785280	0.047295	0.285146	0.055665	0.060464	0.336913	0.068236	0.021854	0.061402	0.063024
785281	0.046069	0.283838	0.054845	0.057951	0.342521	0.068038	0.021745	0.060809	0.064184
785282	0.047428	0.290018	0.054453	0.057945	0.336572	0.067724	0.021447	0.059859	0.064554
785283	0.048316	0.289487	0.054065	0.058682	0.334200	0.070215	0.020822	0.059541	0.064672
785284	0.048527	0.292924	0.056787	0.057575	0.331831	0.070127	0.021034	0.057821	0.063374