Divide two dataframes with python

Question

I have two dataframes : df1 and df2

df1:

TIMESTAMP           eq1 eq2 eq3
2016-05-10 13:20:00  40  30  10
2016-05-10 13:40:00  40  10  20

df2:

TIMESTAMP           eq1 eq2 eq3
2016-05-10 13:20:00  10  20  30
2016-05-10 13:40:00  10  20  20

I would like to divide df1 by df2 : each column of df1 by all column of df2 to get this result df3 :

TIMESTAMP           eq1        eq2        eq3
2016-05-10 13:20:00  40/(10+10) 30/(20+20) 10/(30+20)
2016-05-10 13:40:00  40/(10+10) 10/(20+20) 20/(30+20)

Any idea please?

jezrael · Accepted Answer · 2016-09-12 19:10:18Z

9

You can use div, but before set_index from both columns TIMESTAMP:

df1.set_index('TIMESTAMP', inplace=True)
df2.set_index('TIMESTAMP', inplace=True)

print (df1.div(df2).reset_index())
            TIMESTAMP  eq1  eq2       eq3
0 2016-05-10 13:20:00  4.0  1.5  0.333333
1 2016-05-10 13:40:00  4.0  0.5  1.000000

EDIT by comment:

df1.set_index('TIMESTAMP', inplace=True)
df2.set_index('TIMESTAMP', inplace=True)
print (df2.sum())
eq1    20
eq2    40
eq3    50
dtype: int64

print (df1.div(df2.sum()).reset_index())
            TIMESTAMP  eq1   eq2  eq3
0 2016-05-10 13:20:00  2.0  0.75  0.2
1 2016-05-10 13:40:00  2.0  0.25  0.4

edited Sep 12, 2016 at 19:10

answered Sep 12, 2016 at 16:34

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Alexander Over a year ago

Yes, I had assumed TIMESTAMP was the index.

jezrael Over a year ago

Yes, if it first columns are indexes, solution is simplier df3 = df1.div(df2)

Poisson Over a year ago

@jezrael thak you for your reply, but I made an error, can you please check my post edit? Thank you

Alexander · Accepted Answer · 2016-09-12 19:13:04Z

3

This should work if TIMESTAMP is not the index:

>>> df1.set_index('TIMESTAMP').div(df2.set_index('TIMESTAMP').sum()) 
                     eq1   eq2  eq3
TIMESTAMP                          
2016-05-10 13:20:00    2  0.75  0.2
2016-05-10 13:40:00    2  0.25  0.4

If TIMESTAMP is the index, then simply this:

df1.div(df2.sum())

edited Sep 12, 2016 at 19:13

answered Sep 12, 2016 at 16:17

Alexander

111k32 gold badges212 silver badges208 bronze badges

1 Comment

Poisson Over a year ago

Thank you Alexander , but I made an error, can you please check my post edit? Thank you

Collectives™ on Stack Overflow

Divide two dataframes with python

2 Answers 2

3 Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related