

In [2]:

    
import pandas as pd
from glob import glob
import os
from collections import defaultdict as dd



In [5]:

    
df = pd.read_csv('meanval/wmt15.csv', sep="\t")
df.head()









    Out[5]:






  
    
      
      judgeid
      rank
      rankingid
      segmentid
      srcindex
      srclang
      system
      trglang
      reference
      transition
    
  
  
    
      0
      judge47
      1
      487
      1387
      1387
      ces
      newstest2015.uedin-syntax.4004.cs-en.txt
      eng
      18 machine got loose coming out of turn 2 and ...
      18 got out of control, out of the second bend ...
    
    
      1
      judge47
      2
      487
      1387
      1387
      ces
      newstest2015.CU-TectoMT.4138.cs-en.txt
      eng
      18 machine got loose coming out of turn 2 and ...
      18 it got out of a control, it came from the s...
    
    
      2
      judge47
      4
      488
      2061
      2061
      ces
      newstest2015.uedin-jhu-phrase.4071.cs-en.txt
      eng
      Among the rejected items were the writings of ...
      Among the items we find odmítnutými writings a...
    
    
      3
      judge47
      1
      488
      2061
      2061
      ces
      newstest2015.online-A.0.cs-en.txt
      eng
      Among the rejected items were the writings of ...
      Among the rejected items we find the writings ...
    
    
      4
      judge47
      2
      488
      2061
      2061
      ces
      newstest2015.online-B.0.cs-en.txt
      eng
      Among the rejected items were the writings of ...
      Declined items can be found among the writings...

	judgeid	rank	rankingid	segmentid	srcindex	srclang	system	trglang	reference	transition
0	judge47	1	487	1387	1387	ces	newstest2015.uedin-syntax.4004.cs-en.txt	eng	18 machine got loose coming out of turn 2 and ...	18 got out of control, out of the second bend ...
1	judge47	2	487	1387	1387	ces	newstest2015.CU-TectoMT.4138.cs-en.txt	eng	18 machine got loose coming out of turn 2 and ...	18 it got out of a control, it came from the s...
2	judge47	4	488	2061	2061	ces	newstest2015.uedin-jhu-phrase.4071.cs-en.txt	eng	Among the rejected items were the writings of ...	Among the items we find odmítnutými writings a...
3	judge47	1	488	2061	2061	ces	newstest2015.online-A.0.cs-en.txt	eng	Among the rejected items were the writings of ...	Among the rejected items we find the writings ...
4	judge47	2	488	2061	2061	ces	newstest2015.online-B.0.cs-en.txt	eng	Among the rejected items were the writings of ...	Declined items can be found among the writings...