In [5]:
cd ../report/


/home/scott/Documents/git/neukrill-net-work/report

In [6]:
import csv

In [7]:
import numpy as np

In [8]:
with open('leaderboardscores.csv', 'rb') as csvfile:
    lb = csv.reader(csvfile, delimiter=';')
    for row in lb:
        print row[3]


Score
0.565971
0.580300
0.587967
0.604112
0.606921
0.607333
0.610072
0.610964
0.613981
0.617123
0.619674
0.619722
0.620708
0.623529
0.625620
0.627582
0.629008
0.633647
0.634813
0.637492
0.637510
0.639206
0.641510
0.644150
0.645201
0.645573
0.646626
0.649622
0.649712
0.650670
0.654808
0.655795
0.656547
0.657061
0.660353
0.662100
0.662223
0.662241
0.662924
0.665556
0.667073
0.675780
0.677713
0.677916
0.678478
0.679511
0.681684
0.683645
0.687029
0.689614
0.692339
0.697469
0.699254
0.699467
0.700180
0.701704
0.704582
0.704775
0.704782
0.705380
0.705742
0.706741
0.707233
0.707695
0.713836
0.714217
0.718674
0.719978
0.720527
0.720987
0.722385
0.723187
0.724232
0.725161
0.725254
0.725418
0.728190
0.730720
0.732012
0.732933
0.733551
0.735462
0.738978
0.740716
0.740879
0.741624
0.743406
0.744238
0.745370
0.748310
0.749094
0.750231
0.750697
0.750743
0.751415
0.752094
0.752475
0.752682
0.752869
0.753087
0.755114
0.755295
0.759122
0.762580
0.764690
0.765938
0.766397
0.768097
0.769755
0.769772
0.770842
0.771236
0.771482
0.772689
0.773826
0.774901
0.775228
0.775291
0.775889
0.776284
0.776784
0.776868
0.777557
0.778229
0.778386
0.780433
0.782181
0.784492
0.785569
0.787939
0.792364
0.792476
0.792636
0.793735
0.796267
0.797512
0.801297
0.803575
0.804115
0.804517
0.812123
0.813159
0.813314
0.813716
0.814870
0.815904
0.816074
0.817114
0.820304
0.821521
0.823529
0.823540
0.825269
0.826458
0.826782
0.827810
0.831297
0.832157
0.834590
0.836179
0.837303
0.838255
0.840356
0.840964
0.843820
0.844198
0.844542
0.847867
0.850158
0.854385
0.855241
0.858475
0.861677
0.863606
0.865116
0.867833
0.871236
0.873478
0.874600
0.875768
0.876052
0.881509
0.882130
0.883367
0.888328
0.890085
0.893271
0.896605
0.898542
0.898886
0.900712
0.901230
0.902273
0.903767
0.904069
0.904137
0.905352
0.908481
0.910012
0.913130
0.916023
0.919950
0.920408
0.921474
0.922169
0.923645
0.924679
0.927448
0.935237
0.935645
0.936840
0.938659
0.940522
0.942227
0.943239
0.951936
0.952637
0.952850
0.954263
0.955163
0.956687
0.961772
0.963405
0.965503
0.965960
0.966742
0.968010
0.970017
0.971538
0.971570
0.971942
0.973350
0.974608
0.974850
0.976416
0.976585
0.977135
0.977396
0.977900
0.978967
0.979025
0.984750
0.985966
0.987493
0.987493
0.987815
0.992666
0.996323
0.997694
1.001131
1.009506
1.009699
1.011450
1.016647
1.018010
1.026383
1.031717
1.033327
1.035482
1.035992
1.041526
1.044998
1.045839
1.046305
1.049833
1.050055
1.050766
1.059868
1.071417
1.072072
1.074380
1.076766
1.083163
1.084866
1.088833
1.089005
1.092200
1.096055
1.098720
1.101912
1.103008
1.108609
1.108734
1.112911
1.113133
1.115640
1.116943
1.124535
1.128475
1.132911
1.141507
1.149471
1.157169
1.161414
1.161673
1.161783
1.167266
1.168983
1.176729
1.181642
1.184776
1.186738
1.189349
1.190717
1.190953
1.193860
1.197484
1.201588
1.203163
1.203798
1.208288
1.211943
1.217267
1.222136
1.222351
1.222566
1.228215
1.228503
1.229747
1.233844
1.241520
1.243074
1.250195
1.253741
1.274862
1.275341
1.276468
1.277166
1.278179
1.281175
1.285861
1.289216
1.294394
1.295379
1.299149
1.303136
1.315021
1.319577
1.320512
1.322939
1.332551
1.339141
1.339497
1.342843
1.351688
1.358359
1.365143
1.367245
1.368226
1.371158
1.373435
1.376007
1.377206
1.381451
1.381537
1.382370
1.384112
1.384497
1.384686
1.384746
1.386740
1.387958
1.389328
1.394127
1.394484
1.396032
1.396052
1.397022
1.397984
1.399129
1.399583
1.399703
1.401619
1.401825
1.402172
1.402999
1.403287
1.403574
1.404131
1.404287
1.407560
1.407777
1.410581
1.411682
1.412518
1.413791
1.414461
1.415499
1.415994
1.416676
1.418837
1.421881
1.422699
1.425496
1.425691
1.427430
1.427625
1.432849
1.436124
1.436455
1.436586
1.436655
1.439741
1.441629
1.458145
1.459299
1.461734
1.466391
1.471224
1.472479
1.477400
1.480054
1.488954
1.489425
1.490504
1.497918
1.497988
1.499882
1.505522
1.507244
1.512550
1.520961
1.521299
1.526712
1.527532
1.531933
1.534719
1.547803
1.553338
1.565001
1.574728
1.583230
1.585528
1.587075
1.588981
1.598490
1.616778
1.647741
1.652255
1.653273
1.653358
1.656898
1.661566
1.664645
1.665968
1.669426
1.679515
1.687886
1.701276
1.704748
1.705541
1.707644
1.737042
1.748216
1.760753
1.762166
1.768019
1.772463
1.784778
1.798854
1.810620
1.811868
1.813165
1.814893
1.822912
1.827721
1.829074
1.837086
1.842237
1.847670
1.850615
1.854535
1.856318
1.865422
1.871325
1.878938
1.890211
1.897622
1.901130
1.907950
1.913843
1.924661
1.925373
1.927763
1.929314
1.939973
1.951007
1.951604
1.952125
1.952207
1.957005
1.960133
1.964753
1.974312
1.984107
1.990375
1.991057
2.011439
2.014052
2.017504
2.025091
2.030986
2.033871
2.036350
2.048687
2.051488
2.079291
2.080759
2.085366
2.094338
2.094790
2.107561
2.117703
2.126454
2.142335
2.152715
2.153389
2.157474
2.163301
2.167550
2.173551
2.173581
2.185156
2.196391
2.197002
2.214292
2.218532
2.235007
2.245985
2.246845
2.248132
2.269167
2.269871
2.275809
2.276004
2.278787
2.279902
2.280049
2.293068
2.303918
2.305248
2.308803
2.321829
2.323347
2.323373
2.325202
2.328098
2.332684
2.342751
2.366938
2.368950
2.369048
2.369766
2.372362
2.377840
2.379998
2.383382
2.383851
2.385897
2.390804
2.407486
2.419961
2.427642
2.434292
2.434467
2.438705
2.441377
2.442057
2.448713
2.463525
2.473241
2.474501
2.481733
2.483032
2.484866
2.486346
2.493089
2.499725
2.503599
2.512334
2.519102
2.550234
2.551539
2.553976
2.554533
2.555366
2.564483
2.569068
2.579418
2.583740
2.588582
2.593524
2.620194
2.620719
2.626746
2.637930
2.653109
2.657105
2.668967
2.679293
2.682544
2.687989
2.691352
2.692815
2.706986
2.719213
2.720342
2.721933
2.773131
2.803589
2.822194
2.822709
2.824125
2.824876
2.826727
2.838668
2.853793
2.866185
2.866457
2.882361
2.884034
2.887835
2.900000
2.900978
2.905170
2.910367
2.913083
2.923557
2.926808
2.928097
2.963693
2.969454
3.005369
3.005902
3.012048
3.016048
3.025236
3.034921
3.039154
3.050716
3.085989
3.103884
3.104722
3.105332
3.117419
3.123209
3.125729
3.144448
3.146239
3.150536
3.158541
3.162619
3.201931
3.262386
3.264571
3.266081
3.275882
3.286341
3.288693
3.288775
3.305361
3.326039
3.341117
3.359510
3.411141
3.411943
3.416136
3.418018
3.422119
3.422421
3.422503
3.422759
3.423243
3.423466
3.423862
3.423966
3.424125
3.424433
3.424502
3.424550
3.424843
3.425009
3.425098
3.425397
3.425397
3.425397
3.425397
3.425397
3.425397
3.425397
3.425540
3.425597
3.425936
3.426149
3.426296
3.426723
3.427598
3.446589
3.453585
3.474923
3.497342
3.528190
3.542735
3.564871
3.582036
3.607051
3.620839
3.628628
3.629904
3.653638
3.653664
3.661183
3.677357
3.678106
3.686628
3.695598
3.696134
3.707371
3.712441
3.731752
3.732046
3.736073
3.742662
3.744410
3.748770
3.758068
3.771637
3.772359
3.787888
3.798983
3.814621
3.815713
3.820406
3.822267
3.826216
3.841860
3.872299
3.887454
3.890371
3.942393
4.042344
4.093341
4.103614
4.124615
4.142381
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.161789
4.162038
4.162195
4.172874
4.176986
4.178749
4.189711
4.224316
4.245005
4.287871
4.304033
4.307981
4.332270
4.332561
4.332700
4.336006
4.340318
4.341123
4.347023
4.368273
4.371678
4.409168
4.425855
4.440180
4.473907
4.493356
4.494972
4.499391
4.500724
4.505164
4.570783
4.576048
4.600158
4.613431
4.634807
4.694596
4.706106
4.713458
4.714704
4.716169
4.733144
4.733294
4.738197
4.761677
4.780543
4.782961
4.793395
4.794063
4.794443
4.795774
4.795786
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.795791
4.797186
4.798613
4.798761
4.837443
4.911969
4.912388
5.022295
5.080894
5.095313
5.106396
5.106506
5.121342
5.272354
5.280247
5.322231
5.362395
5.392852
5.775806
5.984996
6.203733
6.352624
6.391673
6.558707
6.711469
6.830148
6.969978
7.769392
7.932273
8.168250
8.410127
9.631407
11.039150
11.244433
14.547863
15.007985
15.157149
15.737932
16.256947
18.074931
18.215245
19.113193
21.748155
21.766870
22.158432
25.527604
29.434228
32.409577
34.520945

In [9]:
import pandas as pd

In [13]:
df = pd.read_csv('leaderboardscores.csv', sep=';')
saved_column = df.Score

In [11]:
print df


         # Δrank  \
0        1     —   
1        2     —   
2        3     —   
3        4    ↑2   
4        5     —   
5        6    ↓2   
6        7    ↑1   
7        8    ↓1   
8        9     —   
9       10    ↑2   
10      11    ↑2   
11      12    ↓2   
12      13    ↓2   
13      14    ↑1   
14      15    ↑2   
15      16     —   
16      17    ↓3   
17      18    ↑9   
18      19    ↑7   
19      20    ↓2   
20      21    ↓2   
21      22    ↓2   
22      23    ↓2   
23      24    ↓1   
24      25    ↓3   
25      26    ↓1   
26      27    ↓3   
27      28     —   
28      29    ↑2   
29      30    ↓1   
...    ...   ...   
1020  1020     —   
1021  1021     —   
1022  1022     —   
1023  1023     —   
1024  1024     —   
1025  1025    ↑1   
1026  1026    ↓1   
1027  1027     —   
1028  1028     —   
1029  1029     —   
1030  1030     —   
1031  1031     —   
1032  1032     —   
1033  1033     —   
1034  1034     —   
1035  1035     —   
1036  1036    ↑1   
1037  1037    ↓1   
1038  1038    ↑1   
1039  1039    ↓1   
1040  1040     —   
1041  1041     —   
1042  1042     —   
1043  1043     —   
1044  1044     —   
1045  1045     —   
1046  1046     —   
1047  1047     —   
1048  1048     —   
1049  1049     —   

     Team Name                                                              * in the money  \
0     ≋ Deep Sea ≋                              ...                                      
1     Happy Lantern Festival                        ...                                      
2     Poisson Process                               ...                                      
3                                               Junonia                                      
4     ⚓Deepsea Challenger⚓                      ...                                      
5                                             AuroraXie                                      
6                                         Maxim Milakov                                      
7                                        Ilya Kostrikov                                      
8                                               old-ufo                                      
9                                              nagadomi                                      
10                                               zzspar                                      
11    Biolab                                        ...                                      
12              Alexander Ryzhkov (MSU, Moscow, Russia)                                      
13                                                beile                                      
14                                              harkmug                                      
15    ‚, Griffin Liang and 2 others last submitted o...                                      
16                                            DeepOcean                                      
17    BAH_kungfu-cats                               ...                                      
18    🐙 axe 🐬 kw 🐙                                  ...                                      
19                                                Pikqu                                      
20                                              toshi_k                                      
21                                         Jae Hyun Lim                                      
22    Apollo                                        ...                                      
23    3 cobblers                                    ...                                      
24                                                 khyh                                      
25    Niklas Köhler                                 ...                                      
26    Matt & Eben                                   ...                                      
27                                          Jianmin Sun                                      
28    Free as air and water                         ...                                      
29                                             L Miksys                                      
...                                                 ...                                      
1020                                             Herman                                      
1021                                       Jason Forbes                                      
1022                                         Andrew Koe                                      
1023  planktonet                                    ...                                      
1024                                       Dennis Liang                                      
1025                                             KeithJ                                      
1026                                          An jinwon                                      
1027                                  Heavy Sweet Crude                                      
1028                                          Jay Moore                                      
1029                                         afterhours                                      
1030                                            Xiaoman                                      
1031                                    Steven Mortimer                                      
1032                                          Jiangyuan                                      
1033                                            BigDota                                      
1034                                            hkhanal                                      
1035                                      pyunpyun_maru                                      
1036                                      SergeySorokin                                      
1037                                           Shturman                                      
1038                                             alex 2                                      
1039                                           onur BAÞ                                      
1040                                      yuanqj9290609                                      
1041                                              r2007                                      
1042                                   Chamara Anuranga                                      
1043                                              Jonas                                      
1044                                      Shayn Weidner                                      
1045                                             Thirst                                      
1046                                           CharlieW                                      
1047                                     scienceguy22 2                                      
1048                                       Daniel Otero                                      
1049                                                bbk                                      

          Score  Entries       Last Submission UTC (Best − Last Submission)  
0      0.565971       96                          Mon, 16 Mar 2015 17:32:03  
1      0.580300      150  Mon, 16 Mar 2015 14:03:40                     ...  
2      0.587967      134  Mon, 16 Mar 2015 23:54:04                     ...  
3      0.604112       59  Mon, 16 Mar 2015 20:24:47                     ...  
4      0.606921      142  Mon, 16 Mar 2015 19:47:18                     ...  
5      0.607333       32                          Mon, 16 Mar 2015 22:15:07  
6      0.610072       27  Sat, 14 Mar 2015 18:36:14                     ...  
7      0.610964       51  Mon, 16 Mar 2015 20:47:09                     ...  
8      0.613981       82                          Mon, 16 Mar 2015 21:23:27  
9      0.617123       31  Mon, 16 Mar 2015 05:57:34                     ...  
10     0.619674      174  Mon, 16 Mar 2015 18:47:59                     ...  
11     0.619722       74  Mon, 16 Mar 2015 14:50:21                     ...  
12     0.620708       67  Mon, 16 Mar 2015 23:26:23                     ...  
13     0.623529       28  Mon, 16 Mar 2015 21:01:15                     ...  
14     0.625620       28  Mon, 16 Mar 2015 20:42:44                     ...  
15     0.627582      168  Mon, 16 Mar 2015 02:37:25                     ...  
16     0.629008       64                          Mon, 16 Mar 2015 20:30:26  
17     0.633647       49                          Mon, 16 Mar 2015 23:42:59  
18     0.634813      137  Mon, 16 Mar 2015 18:09:35                     ...  
19     0.637492       61  Mon, 16 Mar 2015 13:27:53                     ...  
20     0.637510      120  Mon, 16 Mar 2015 20:35:07                     ...  
21     0.639206       46  Mon, 16 Mar 2015 16:13:49                     ...  
22     0.641510      242  Mon, 16 Mar 2015 23:54:24                     ...  
23     0.644150      111  Mon, 16 Mar 2015 17:20:36                     ...  
24     0.645201       36                          Mon, 16 Mar 2015 23:42:42  
25     0.645573       24                          Thu, 12 Mar 2015 10:49:48  
26     0.646626       42  Mon, 16 Mar 2015 23:36:42                     ...  
27     0.649622       67  Mon, 23 Feb 2015 23:33:30                     ...  
28     0.649712       75  Mon, 16 Mar 2015 22:07:24                     ...  
29     0.650670       32  Mon, 16 Mar 2015 12:49:14                     ...  
...         ...      ...                                                ...  
1020   5.984996        1                          Mon, 09 Mar 2015 18:47:39  
1021   6.203733        3                          Sat, 14 Feb 2015 19:36:32  
1022   6.352624        2                          Sat, 14 Feb 2015 15:34:51  
1023   6.391673        6  Tue, 03 Mar 2015 08:38:04                     ...  
1024   6.558707        4                          Sat, 14 Feb 2015 14:48:32  
1025   6.711469        1                          Thu, 05 Mar 2015 08:49:44  
1026   6.830148        3  Tue, 24 Feb 2015 06:50:50                     ...  
1027   6.969978        1                          Tue, 27 Jan 2015 01:01:00  
1028   7.769392        1                          Fri, 02 Jan 2015 06:29:29  
1029   7.932273        2  Tue, 10 Feb 2015 04:25:48                     ...  
1030   8.168250        1                          Tue, 06 Jan 2015 08:38:55  
1031   8.410127        1                          Wed, 17 Dec 2014 13:47:41  
1032   9.631407        1  Sun, 04 Jan 2015 04:30:33                     ...  
1033  11.039150        3  Fri, 20 Feb 2015 08:23:12                     ...  
1034  11.244433        2                          Tue, 06 Jan 2015 04:42:38  
1035  14.547863        1                          Mon, 09 Mar 2015 23:00:55  
1036  15.007985        4                          Sun, 25 Jan 2015 09:44:17  
1037  15.157149        5  Thu, 12 Feb 2015 10:55:00                     ...  
1038  15.737932        1                          Sun, 18 Jan 2015 19:56:24  
1039  16.256947        4  Wed, 11 Mar 2015 19:21:52                     ...  
1040  18.074931        1                          Wed, 11 Feb 2015 15:43:50  
1041  18.215245        3                          Wed, 21 Jan 2015 13:49:24  
1042  19.113193        1                          Wed, 14 Jan 2015 15:21:43  
1043  21.748155        5                          Thu, 05 Feb 2015 13:51:17  
1044  21.766870        1                          Tue, 03 Mar 2015 07:16:13  
1045  22.158432        1                          Mon, 29 Dec 2014 17:52:41  
1046  25.527604        1                          Tue, 03 Mar 2015 22:54:21  
1047  29.434228        1                          Sun, 04 Jan 2015 09:02:16  
1048  32.409577        2  Mon, 09 Feb 2015 15:10:19                     ...  
1049  34.520945        1                          Mon, 09 Mar 2015 17:41:23  

[1050 rows x 6 columns]

In [14]:
scores = []
with open('leaderboardscores.csv', 'rb') as csvfile:
    lb = csv.reader(csvfile, delimiter=';')
    for i,row in enumerate(lb):
        if i==0:
            continue
        scores.append(float(row[3]))

In [15]:
scores = np.array(scores)

In [16]:
scores


Out[16]:
array([  0.565971,   0.5803  ,   0.587967, ...,  29.434228,  32.409577,
        34.520945])

In [17]:
import matplotlib.mlab as mlab
import matplotlib.pyplot as plt

In [18]:
%pylab
%matplotlib inline


Using matplotlib backend: agg
Populating the interactive namespace from numpy and matplotlib

In [19]:
# evaluate the histogram
values, base = np.histogram(scores, bins=len(np.unique(scores)))
#evaluate the cumulative
cumulative = np.cumsum(values)
cumulative = (cumulative+0.0) / cumulative[-1]
# plot the cumulative function
plt.plot(base[:-1], cumulative, c='blue')
plt.plot([scores[0],scores[0]],[0,1], c='green')
plt.plot([0.704582,0.704582],[0,1], c='green')

plt.xlabel('Leaderboard log-loss')
plt.ylabel('CDF')

plt.grid(True)
plt.show()



In [20]:
# evaluate the histogram
values, base = np.histogram(scores, bins=250)

In [21]:
li = scores<4.795791

# evaluate the histogram
values, base = np.histogram(scores[li], bins=100)

# plot the cumulative function
#plt.plot(base[:-1], values, c='blue')
plt.bar(base[:-1], values, width=(base[1]-base[0]))

plt.plot([scores[0],scores[0]],[0,45], c='green')
plt.plot([0.704582,0.704582],[0,45], c='red')
plt.plot([4.161789,4.161789],[0,45], c='grey')
#plt.plot([4.795791,4.795791],[0,45], c='grey')

plt.xlabel('Leaderboard log-loss')
plt.ylabel('Count')

plt.grid(True)

plt.savefig("leaderboard.svg", close=False, verbose=True)



In [22]:
import seaborn as sns

In [65]:
li = scores<4.795791

# evaluate the histogram
values, base = np.histogram(scores[li], bins=100)

# plot the cumulative function
#plt.plot(base[:-1], values, c='blue')
plt.bar(base[:-1], values, width=(base[1]-base[0]), edgecolor='None')

plt.plot([scores[0],scores[0]],[0,45], '--', c='green')
plt.plot([0.704582,0.704582],[0,45], '--', c=[1, .25, .25])
plt.plot([4.161789,4.161789],[0,45], '--', c=[.25, .25, .25])
plt.plot([4.795791,4.795791],[0,45], '--', c='black')

plt.tick_params(axis='both', which='major', labelsize=22)
plt.tick_params(axis='both', which='minor', labelsize=22)

plt.xlabel('Log-loss score on leaderboard', fontsize=32)
plt.ylabel('Number of teams', fontsize=32)

plt.grid(True)

plt.savefig("leaderboard_sb.svg", bbox_inches='tight')