Can I write an array to a Panda DataFrame row?

Question

Hi I have a table like this

id q         
1  hello?     
2  helloWorld 
3  hi

I want to create another column with name 'k' like this,

id q              k
1  hello?        [h,e,l,o]
2  helloWorld    [h,e,l,o,w,r,d]
3  hi            [h, i]

The type of k row is nparray, is this possible with Pandas?

if you already have the rows of k stored somehow, like in a variable kData, then you'd just do existing_dataframe['k'] = kData. However, it's unclear what you have in memory and what you want to do. Are you generating the rows of k from the existing dataframe? — shortorian
– shortorian, Commented May 22, 2021 at 1:10
I have a kValue = [h, e, l, o] , I want to insert something like existing_dataframe['k'][0] = kValue — T PHo
– T PHo, Commented May 22, 2021 at 1:18
Then your kValues should be included in the body of your question. — Henry Ecker
– Henry Ecker ♦, Commented May 22, 2021 at 1:19
I think I was able to do this somehow, thanks for the answers, though it is helpful for me — T PHo
– T PHo, Commented May 22, 2021 at 1:21
it is just an example, I thought panda rows doesn't accept array values, but apparently it do — T PHo
– T PHo, Commented May 22, 2021 at 1:22

Henry Ecker · Accepted Answer · 2021-05-22 01:17:31Z

3

Try filtering out non-alphabetical characters with replace then apply + pd.unqiue:

import pandas as pd

df = pd.DataFrame({
    'id': [1, 2, 3],
    'q': ['hello?', 'helloWorld', 'hi']
})

df['k'] = df['q'].replace(r'\W', '', regex=True) \
    .apply(lambda x: pd.unique(list(x.lower())))

print(df)

df:

   id           q                      k
0   1      hello?           [h, e, l, o]
1   2  helloWorld  [h, e, l, o, w, r, d]
2   3          hi                 [h, i]

Or if order doesn't matter set is an option:

import numpy as np
import pandas as pd

df = pd.DataFrame({
    'id': [1, 2, 3],
    'q': ['hello?', 'helloWorld', 'hi']
})

df['k'] = df['q'].replace(r'\W', '', regex=True) \
    .apply(lambda x: np.array([*set(x.lower())]))

print(df)

df:

   id           q                      k
0   1      hello?           [h, o, l, e]
1   2  helloWorld  [o, e, r, d, h, l, w]
2   3          hi                 [i, h]

edited May 22, 2021 at 1:17

answered May 22, 2021 at 1:11

Henry Ecker♦

35.8k19 gold badges48 silver badges67 bronze badges

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

Collectives™ on Stack Overflow

Can I write an array to a Panda DataFrame row?

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related