How to find most frequnet words in a corpus in Pandas dataframe (Python)

Question

I have Pandas dataframe that looks like following.I have tokenized my text files and used NLTK Countvectorizer to convert into pandas dataframe. In addition, I have already removed stopwords and punctuation from my coupus. I am trying to find most frequent words in my corpus in pandas dataframe. In below dataframe,words such as "aaron" and "abandon" aprreared >10 times, thus those words should be in new dataframe.

Note: I am new to python, and I am not sure how to implement this. Provide explanation with code.

Subset of the dataframe

I already already clean my corpus and my dataframe looks like following

{'aaaahhhs': {990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 0, 997: 0, 998: 0, 999: 0, 1000: 1}, 'aahs': {990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 0, 997: 0, 998: 0, 999: 0, 1000: 1}, 'aamir': {990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 0, 997: 0, 998: 0, 999: 0, 1000: 1}, 'aardman': {990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 0, 997: 0, 998: 0, 999: 0, 1000: 2}, 'aaron': {990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 4, 997: 0, 998: 0, 999: 0, 1000: 14}, 'abandon': {990: 0, 991: 0, 992: 0, 993: 0, 994: 0, 995: 0, 996: 0, 997: 0, 998: 0, 999: 0, 1000: 16}}

Answer 1

If need top N words:

N = 2 
print (df.sum().nlargest(N).index)
Index(['aaron', 'abandon'], dtype='object')

Another solution:

print (df.sum().sort_values(ascending=False).index[:N])
Index(['aaron', 'abandon'], dtype='object')

If need also counts in one column DataFrame or Series (remove to_frame):

N = 2
print (df.sum().nlargest(N).to_frame('count'))
         count
aaron       18
abandon     16
print (df.sum().sort_values(ascending=False).iloc[:N].to_frame('count'))
         count
aaron       18
abandon     16

If need 2 column DataFrame:

print (df.sum().nlargest(N).rename_axis('word').reset_index(name='count'))
      word  count
0    aaron     18
1  abandon     16

print (df.sum()
         .sort_values(ascending=False).iloc[:N]
         .rename_axis('word')
         .reset_index(name='count'))
      word  count
0    aaron     18
1  abandon     16