import numpy as np
import pandas as pd
import seaborn as sn
from scipy import stats
#import missingno as msno
from datetime import datetime
import matplotlib.pyplot as plt
import os
import sys
import codecs
import nltk
from nltk.corpus import stopwords
from sklearn.feature_extraction.text import CountVectorizer

default_stopwords = set(nltk.corpus.stopwords.words('english'))

def get_words(df):
    return (' '.join(df['description']))

data_df = pd.read_json("../data/train.json")
wordbag = get_words(data_df)

words = nltk.word_tokenize(


vec = CountVectorizer(analyzer=u'word', stop_words="english").fit([wordbag])
print vec