final.py

import scipy.io
from scipy.sparse import coo_matrix
import json,os,time,math,operator,enchant,string
import numpy as np
from nltk.corpus import stopwords
words=enchant.Dict("en_US")
stop = stopwords.words('english') + list(string.punctuation)

def preprocess_text(arg):
	arg=[k for k in arg if k.isalnum() or k==" " or k=="\n"]
	arg=[word for word in arg if word not in stop and words.check(word)]
	arg=[k.lower() for k in arg]
	arg=[k for k in voc]
	return arg
if __name__ == '__main__':
	train_js={}
	dt=scipy.io.mmread(os.getcwd()+r'/input/jsons/'+"dtm") #loading matrix
	dtm=dt.toarray()
	#loading trained values
	try:
		with open(os.getcwd()+r'/input/jsons/'+"trained",'r') as file1:
			dp=file1.read()
			file1.close()
		train_js=json.loads(dp)
	except IOError as err:
		print err
	cat_list=train_js['cat_list']
	cat_prob=train_js['cat_prob']
	cat_list_value=train_js['cat_list_value']
	vocab_value=train_js['vocab_value']
	voc=train_js['vocab']
	
	sent=preprocess_text("Chinese Chinese Chinese Tokyo Japan")
	proab={}
	for keys in cat_prob:
		proab[keys]=0.0
	for i in xrange(len(sent)):
		sent[i]=sent[i].lower()
	for key in cat_prob: #for a particluar category
		cat=key #find the category name
		cat_present_doc=cat_list_value[cat] #find the docs present in the categories
		for word in sent: #for each word
			words_in_class=0 #to find the vocab of a class
			word_freq_class=0
			vocab_val=vocab_value[word] #find the index of each word in our vocab 
			for i in xrange(len(cat_present_doc)):
				word_freq_class+=dtm[cat_present_doc[i],vocab_val]
				words_in_class+=np.sum(dtm[cat_present_doc[i]])
			word_freq_class+=1			
			proab[cat]+=math.log(float(word_freq_class)/(float(words_in_class)+len(voc))) #to calculate the probablity by computing its log instead and adding 
		proab[cat]+=math.log(float(cat_prob[cat]))#adding naives posterior probablity of the category
	sorted_proab = sorted(proab.items(), key=operator.itemgetter(1))#sorting the dicitionary by values in ascending value
	print(sorted_proab)
 	print(sorted_proab[len(sorted_proab)-1][0])