1

in a a pandas df (data from a csv file) I am trying to add new columns (ratios) by dividing each column by every other column.

So far I am stuck in the process of dividing all columns by the first one (hoping to iterate the process later).

ratio_df = df.join(df.div(df['Hv10'], axis=0), rsuffix='_new_ratio')

I am getting this code from a post that deals with a similar problem.

pandas dataframe create new columns and fill with calculated values from same df

I am getting the following error message:

ValueError: operands could not be broadcast together with shapes (235170,) (3618,)

I am not sure why I am getting this error message since I am dividing each column by another (so dimension should be the same)

What am I doing wrong?

Is there a one step process to generate all these new ratio columns?

I hope my description is clear.

Thanks!

cs95
  • 379,657
  • 97
  • 704
  • 746
Diego
  • 637
  • 3
  • 10
  • 24

1 Answers1

2

You're on the right track, but a join is not the right operation. You should be able to do this using pd.concat.

pd.concat([df.div(df[col], axis=0) for col in df.columns], axis=1) # each column with every other column

If you want to avoid dividing a column with itself, you could use df.columns.difference:

pd.concat([df[df.columns.difference([col])].div(df[col], axis=0) \
                                       for col in df.columns], axis=1)

You can also use df.add_suffix('_new_ratio') to add suffixes to your columns.

MCVE:

import pandas as pd
import numpy as np

np.random.seed([3, 14])
df = pd.DataFrame(np.random.randn(10, 3), columns=list('ABC'))

df

          A         B         C
0 -0.602923 -0.402655  0.302329
1 -0.524349  0.543843  0.013135
2 -0.326498  1.385076 -0.132454
3 -0.407863  1.302895 -0.604236
4 -0.243362 -0.211261 -2.056621
5  0.517868 -0.040749 -1.051875
6  0.607092 -2.230437 -0.610389
7  0.223345  0.841994 -1.564391
8  0.031653  0.655489 -0.288834
9 -0.467438  0.119117  1.519430

df_new = pd.concat([df[df.columns.difference([col])].div(df[col], axis=0)\
                           .add_suffix('_n_r') for col in df.columns], axis=1)
df_new

       B_n_r     C_n_r      A_n_r      C_n_r      A_n_r      B_n_r
0   0.667838 -0.501438   1.497369  -0.750838  -1.994263  -1.331845
1  -1.037176 -0.025050  -0.964156   0.024152 -39.919620  41.403685
2  -4.242213  0.405682  -0.235726  -0.095630   2.464987 -10.457000
3  -3.194442  1.481468  -0.313044  -0.463764   0.675006  -2.156269
4   0.868095  8.450867   1.151948   9.734958   0.118331   0.102723
5  -0.078686 -2.031166 -12.708707  25.813488  -0.492328   0.038739
6  -3.673971 -1.005432  -0.272185   0.273663  -0.994598   3.654123
7   3.769924 -7.004363   0.265257  -1.857959  -0.142768  -0.538225
8  20.708576 -9.125012   0.048289  -0.440639  -0.109589  -2.269430
9  -0.254830 -3.250547  -3.924192  12.755771  -0.307641   0.078396
cs95
  • 379,657
  • 97
  • 704
  • 746