CODEWITHSUNDEEP
pythonpandasdataframescikit-learncountvectorizer
Nabih Bawazir
Nabih Bawazir

Reputation: 7255

How do I merge data with CountVectorizer features

Here's my dataset

        body                                            customer_id   name
14828   Thank you to apply to us.                       5458          Sender A
23117   Congratulation your application is accepted.    5136          Sender B
23125   Your OTP will expire in 10 minutes.             5136          Sender A

Here's my code

import pandas as pd
from sklearn.feature_extraction.text import CountVectorizer
b = a['body']
vect = CountVectorizer()
vect.fit(b)
X_vect=vect.transform(b)
pd.DataFrame(X_vect.toarray(), columns=vect.get_feature_names())

The output is

    10  application apply ... your  
0   0   0           1         0
1   0   1           0         1
2   1   0           0         1

What I need is

        body                                            customer_id   name        10  application apply ... your
14828   Thank you to apply to us.                       5458          Sender A    0   0           1         0
23117   Congratulation your application is accepted.    5136          Sender B    0   1           0         1
23125   Your OTP will expire in 10 minutes.             5136          Sender A    1   0           0         1

How suppose I do this? I'm still hoping to use CountVectorizer so I can modify the function in the future

Upvotes: 1

Views: 1435

Answers (1)

jezrael
jezrael

Reputation: 862691

You can add index to Dataframe contructor and then join to original df with default left join:

b = pd.DataFrame(X_vect.toarray(), columns=vect.get_feature_names(), index= a.index)
a = a.join(b)

Or use merge, but need more parameters, because default is inner join:

a = a.merge(b, left_index=True, right_index=True, how='left')

Upvotes: 3

Related Questions