1

I have the following dataframe df1

         value
0     0.164560
1     0.000000
2     0.350000
3     0.700000
    ...
3778  0.350000
3779  0.000000
3780  0.137500
3781  0.253333

and another dataframe df2

                 0             1
2669  1.744478e-05  2.323815e-05
5417  2.274561e-06  5.808474e-04
6102  2.220705e-06  1.605110e-04
4033  2.018643e-04  2.867235e-05
4877  4.549740e-07  6.233500e-04
             ...
4470  2.761791e-06  6.504309e-05
5071  1.261781e-06  8.010408e-05
3635  4.810822e-05  3.325764e-05
5378  1.133819e-04  4.934986e-05

by joining the two dataframes I want the index from df2 to be an id column in the joined dataframe by removing columns 0 and 1 from df2. So the desired output looks like

  id         value
2669     0.164560
5417     0.000000
6102     0.350000
4033     0.700000
    ...
4470     0.350000
5071     0.000000
3635     0.137500
5378     0.253333

I tried 

df3 = pd.concat([df2, df1], ignore_index=True)

Output of the joined dataframe with pd.concat

         value             0             1
0     0.164560           NaN           NaN
1     0.000000           NaN           NaN
2     0.350000           NaN           NaN
3     0.700000           NaN           NaN
          ....
7382       NaN  4.165176e-05  2.615312e-05
7383       NaN  2.357132e-05  5.779454e-05
7384       NaN  1.761602e-06  7.960426e-05
7385       NaN  9.162049e-11  1.627316e-04
Samuel Mideksa
  • 423
  • 8
  • 19

1 Answers1

2

If length of both DataFrames are same only assign one index to new column:

df1['id'] = df2.index
print (df1)

Or:

df1 = df1.assign(id = df2.index)
print (df1)
jezrael
  • 822,522
  • 95
  • 1,334
  • 1,252
  • I am getting `ValueError: Length of values does not match length of index` How can I compare the length of the indexes for the dataframes? – Samuel Mideksa Feb 01 '19 at 12:22
  • @SamuelMideksa - Sorry, I need more information. Why you need add index from `df2` to df1 column if different length of data? – jezrael Feb 01 '19 at 12:45
  • I had to remove some rows from df1 in order to make the indexes equal It woks now but instead of adding an 'id' column can I get actually get the 'id' column as index values in df1? – Samuel Mideksa Feb 01 '19 at 12:53
  • https://stackoverflow.com/questions/54687073/how-to-insert-column-name-in-pandas-dataframe please I do need help with that if you take a look at it this question? – Samuel Mideksa Feb 14 '19 at 09:30
  • @SamuelMideksa - Sure added answer. Rather 2 situtation, because not sure if need add 1 to index or to new column. – jezrael Feb 14 '19 at 09:51