def parallel():
query = ["stars", 5]
projection = ["text", 1]
# Params: query, collection name, projection, limit
documents = PyM.fetchFromDb(query, 'reviews', projection, 3000)
starttime = timer()
pool = mp.Pool(processes=4)
results = [
pool.apply_async(createNGram, args=(doc, )) for doc in documents
]
output = [p.get() for p in results]
flat_ngram_list = [item for sublist in output for item in sublist]
endtime = timer()
fdist = nltk.FreqDist(flat_ngram_list)
sorted_x = sorted(fdist.items(), key=operator.itemgetter(1), reverse=True)
for k in sorted_x[:5]:
print " ".join(k[0]) + ":" + str(k[1])
print "Elapsed time " + str(endtime - starttime)
def singleTh():
query = ["stars", 5]
projection = ["text", 1]
# Params: query, collection name, projection, limit
documents = PyM.fetchFromDb(query, 'reviews', projection, 3000)
starttime = timer()
gram = []
for doc in documents:
gram += createNGram(doc)
# print type(gram[0])
endtime = timer()
fdist = nltk.FreqDist(gram)
sorted_x = sorted(fdist.items(), key=operator.itemgetter(1), reverse=True)
for k in sorted_x[:5]:
print " ".join(k[0]) + ":" + str(k[1])
print "Elapsed time " + str(endtime - starttime)
