CODEWITHSUNDEEP
pandasdataframerows
astroluv
astroluv

Reputation: 793

How to get the value of a single row of columns with same name?

I have data frames with same column names so i have merge them

df1

        wave   num   stlines 
0    4050.32   3.0  0.282690
1    4208.98   5.5  0.490580
2    4374.94   9.0  0.714830
3    4379.74   9.0  0.314040
4    4398.01  14.0  0.504150
5    4502.21   8.0  0.562780

df2

        wave  num  stlines 
0    4050.32    3  0.28616 
1    4208.98    6  0.48781 
2    4374.94    9  0.71548 
3    4379.74   10  0.31338 
4    4398.01   15  0.49950  
5    4502.21    9  0.56362

df3

        wave   num   stlines 
0    4050.32   3.0  0.282690
1    4208.98   7.5  0.490580
2    4374.94   9.0  0.714830
3    4379.74   9.0  0.314040
4    4398.01  14.0  0.504150
5    4502.21   8.0  0.562780

after merging, the resultant dataframe looks like this:

df=pd.merge(df1,df2,df3, on='wave',axis=1,join='inner')


      wave   num_x  stlines_x  num_x  stlines_x  num_x  stlines_x 
0    4050.32   3.0  0.282690    3     0.28616      3.0  0.282690
1    4208.98   5.5  0.490580    6     0.48781      5.5  0.490580 
2    4374.94   9.0  0.714830    9     0.71548      9.0  0.714830
3    4379.74   9.0  0.314040   10     0.31338      9.0  0.314040
4    4398.01  14.0  0.504150   15     0.49950     14.0  0.504150
5    4502.21   8.0  0.562780    9     0.56362      8.0  0.562780

So now if i want to take the values of all the coulmns with namenum_x for any row. Then how can i get them?

I can get the complete columns with same name using the following 

df.num_x

    num   num   num
 0    3    3.0    3
 1  5.5      6  7.5 
 2    9    9.0    9
 3   10   14.0   10 
 4   15    8.0   15
 5    9    3.0    9

but when i tried to do the same for a single row '1' then it didn't work:

df.num_x['1']

The desired result should look like this:

    num   num   num
 1  5.5     6    7.5

How can i get them??

Upvotes: 1

Views: 35

Answers (1)

jezrael
jezrael

Reputation: 863166

You need DataFrame.loc:

df.loc[1, 'num_x']

In pandas same columns names are problematic, because not easy seelct first, second num_x, so suggest create MultiIndex:

dfs = [df1, df2, df3]

df = pd.concat([x.set_index('wave') for x in dfs], 
               axis=1, 
               keys=['df1','df2','df3'], join='inner')
print (df)
          df1          df2            df3         
          num  stlines num  stlines   num  stlines
wave                                              
4050.32   3.0  0.28269   3  0.28616   3.0  0.28269
4208.98   5.5  0.49058   6  0.48781   7.5  0.49058
4374.94   9.0  0.71483   9  0.71548   9.0  0.71483
4379.74   9.0  0.31404  10  0.31338   9.0  0.31404
4398.01  14.0  0.50415  15  0.49950  14.0  0.50415
4502.21   8.0  0.56278   9  0.56362   8.0  0.56278

And then use xs for selecting:

df1 = df.xs('num', axis=1, level=1)
print (df1)
          df1  df2   df3
wave                    
4050.32   3.0    3   3.0
4208.98   5.5    6   7.5
4374.94   9.0    9   9.0
4379.74   9.0   10   9.0
4398.01  14.0   15  14.0
4502.21   8.0    9   8.0

Upvotes: 1

Related Questions