How to modify all values in a pandas dataframe with another numpy array

Question

I have the following example of code

import pandas as pd

df = pd.DataFrame({'a':[1, 2, 3],
                   'b':[10, 20, 30],
                   'c':[100, 200, 300],
                   'd':['q', 'w', 'r']})

Get the values of the dataframe

In [12]: df[['a', 'b', 'c']].values
Out[12]: 
array([[  1,  10, 100],
       [  2,  20, 200],
       [  3,  30, 300]])

Then i normalize the values

from sklearn.preprocessing import normalize

norm = normalize(df[['a', 'b', 'c']].values, axis=0)

In [11]: norm
Out[11]: 
array([[0.26726124, 0.26726124, 0.26726124],
       [0.53452248, 0.53452248, 0.53452248],
       [0.80178373, 0.80178373, 0.80178373]])

Now i want to do something like

df[['a', 'b', 'c']].values = norm

But i get the error (i knew about it)

AttributeError: can't set attribute

How can i modify those values without affecting the other parts of the dataframe (i.e the column 'e' and the indices); just the values.

Thanks.

BENY · Accepted Answer · 2018-08-05 00:51:58Z

3

You do not need call value

df[['a', 'b', 'c']]=norm
df
Out[342]: 
          a         b         c  d
0  0.267261  0.267261  0.267261  q
1  0.534522  0.534522  0.534522  w
2  0.801784  0.801784  0.801784  r

answered Aug 5, 2018 at 0:51

BENY

324k22 gold badges176 silver badges250 bronze badges

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

7 Comments

moctarjallo Over a year ago

Hey now i want to replace column d with something like

moctarjallo Over a year ago

[1, 0, 0], [0, 1, 0], [0, 0, 1] for every category q, w and r. How can I ? Thanks.

BENY Over a year ago

@mctrjallohdf.d=[[1, 0, 0], [0, 1, 0], [0, 0, 1]]

moctarjallo Over a year ago

because i want to make it ready to train a keras model classifier

moctarjallo Over a year ago

I want to compute those binary vectors in the first place

|

Collectives™ on Stack Overflow

How to modify all values in a pandas dataframe with another numpy array

1 Answer 1

7 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

7 Comments

Your Answer

Sign up or log in

Post as a guest

Related