In [8]:

    

from sklearn.feature_extraction.text import TfidfVectorizer
import pandas as pd

# taken from http://stats.stackexchange.com/questions/73908/search-in-tf-idf


    

In [5]:

    

vectorizer = TfidfVectorizer(min_df=1)

my_phrases = ["boring answer phrase",
              "exciting phrase",
              "phrase on stackoverflow",
              "answer on stackoverflow"]

my_features = vectorizer.fit_transform(my_phrases)

feature_words = vectorizer.get_feature_names()


    

In [6]:

    

D = {}
for doc_index, weights in enumerate(my_features.A):
    D[doc_index] = pd.Series(weights, index=feature_words)
df = pd.DataFrame(D)

df_t = df.transpose()


    

In [7]:

    

df_t


    

    
Out[7]:






  

    

      

      
answer
      
boring
      
exciting
      
on
      
phrase
      
stackoverflow
    
  
  

    

      
0
      
0.553492
      
0.702035
      
0.000000
      
0.000000
      
0.448100
      
0.000000
    
    

      
1
      
0.000000
      
0.000000
      
0.842926
      
0.000000
      
0.538029
      
0.000000
    
    

      
2
      
0.000000
      
0.000000
      
0.000000
      
0.613667
      
0.496816
      
0.613667
    
    

      
3
      
0.577350
      
0.000000
      
0.000000
      
0.577350
      
0.000000
      
0.577350