[FEA] Supporting get_feature_names for TfidfVectorizer

[mayankanand007]

Is your feature request related to a problem? Please describe.
I'm looking to get similar functionality from TfidfVectorizer as we have in CountVectorizer in the form of get_feature_names()

Describe the solution you'd like

corpus = ['This is the first document.' , 'This document is the second document.' , 'And this is the third one.' , 
'Is this the first document?']
vectorizer = TfidfVectorizer()
X = vectorizer.fit_transform(corpus)
print(vectorizer.get_feature_names())

# Output: ['and', 'document', 'first', 'is', 'one', 'second', 'the', 'third', 'this']

Describe alternatives you've considered
vec.vocabulary_ is an available attribute to return array mapping from feature integer indices to feature name, however since sklearn has this method available, then it will be great if we can get it in cuML as well.

Additional context
https://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.TfidfVectorizer.html
https://docs.rapids.ai/api/cuml/stable/api.html#cuml.feature_extraction.text.TfidfVectorizer

[mayankanand007]

After looking through the codebase, I realized that since TfidfVectorizer inherits from CountVectorizer class, it also inherits the implementation of .get_feature_names() from there. I fired up a quick example which showed it to work:

from cuml.feature_extraction.text import TfidfVectorizer
import cudf

corpus = ['This is the first document.' , 'This document is the second document.' , 'And this is the third one.' , 
'Is this the first document?']
vectorizer = TfidfVectorizer()
corpus = cudf.Series(corpus)
X = vectorizer.fit_transform(corpus)
print(vectorizer.get_feature_names())

# output: 
# 0         and
# 1    document
# 2       first
# 3          is
# 4         one
# 5      second
# 6         the
# 7       third
# 8        this
Name: token, dtype: object

However, I feel we should still add it to the documentation as well since the functionality is already present (just as sklearn added here)

[github-actions]

This issue has been labeled inactive-30d due to no recent activity in the past 30 days. Please close this issue if no further response or action is needed. Otherwise, please respond with a comment indicating any updates or changes to the original issue and/or confirm this issue still needs to be addressed. This issue will be labeled inactive-90d if there is no activity in the next 60 days.