Array of strings to dataframe with word columns

Question

What's the easiest way to get from an array of strings like this:

arr = ['abc def ghi', 'def jkl xyz', 'abc xyz', 'jkl xyz']

to a dataframe where each column is a single word and each row contains 0 or 1 depending if the word appeared in the string. Something like this:

   abc def ghi jkl xyz
0    1   1   1   0   0
1    0   1   0   1   1
2    1   0   0   0   1
3    0   0   0   1   1

EDIT: Here is my approach, which to me seemed like a lot of python looping and not using the built in pandas functions

labels = (' ').join(arr)
labels = labels.split()
labels = list(set(labels))
labels = sorted(labels)

df = pd.DataFrame(np.zeros((len(arr), len(labels))), columns=labels)
cols = list(df.columns.values)

for i in range(len(arr)):
    for col in cols:
        if col in arr[i]:
            df.set_value(i, col, 1)

I'm sorry, but this site is not meant to solve your tasks, but to help you with problems you occur on your way to solve them. So, what does your code look like so far? — Aiyion.Prime
– Aiyion.Prime, Commented Apr 23, 2017 at 11:15
i included my own code in the question, which works but seemed like a lot of manual python loops. i thought there might be an easier way to do it with pandas — pietz
– pietz, Commented Apr 23, 2017 at 11:43

TLousky · Accepted Answer · 2017-04-23 11:49:42Z

3

EDITED - reduced to 3 essential lines:

import pandas as pd

arr = ['abc def ghi', 'def jkl xyz', 'abc xyz', 'jkl xyz']

words = set( ' '.join( arr ).split() )
rows  = [ { w : int( w in e ) for w in words } for e in arr ]
df    = pd.DataFrame( rows )

print( df )

Result:

   abc  def  ghi  jkl  xyz
0    1    1    1    0    0
1    0    1    0    1    1
2    1    0    0    0    1
3    0    0    0    1    1

edited Apr 23, 2017 at 11:49

answered Apr 23, 2017 at 11:23

TLousky

3032 silver badges8 bronze badges

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

3 Comments

pietz Over a year ago

thanks and sorry for now including my code from the beginning. yours runs a little faster than mine

TLousky Over a year ago

Sure. You can reduce it to 2 lines by creating the DataFrame directly on the result of the rows list comprehension, but this is a bit more readable :)

Parfait Over a year ago

@pietz if this solution helped, please accept it (tick mark to side) to confirm resolution.

Collectives™ on Stack Overflow

Array of strings to dataframe with word columns

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related