def generate_token_vectors(people):
vectors = {}
for ID in people:
vector = generate(people[ID][data])
vectors[ID]['name'] = people[ID]['name']
vectors[ID]['tokens'] = vector
return vectors
def generate_token_vectors(people):
vectors = defaultdict(dict)
for ID in people:
try:
vector = generate(people[ID]['data'])
vectors[ID]['name'] = people[ID]['name']
vectors[ID]['tokens'] = vector.split() # Need a string
except AttributeError:
# Handling the one NoneType blob
pass
return vectors
def queryVector (rawQuery):
cleanQuery = generate(rawQuery)
tokens = cleanQuery.split(' ')
doc = {}
queryTfidf = {}
for token in tokens:
currentVal = doc.get(token, 0)
newVal = currentVal + 1
doc[token] = newVal
for tok in doc.keys():
#if (tok in idfIndex):
#tokenIdf = idfIndex[token]
#tokenTf = doc[token]
#score = tokenIdf * tokenTf
#queryTfidf[tok] = score
tokenIdf = idfIndex.get(token, 0)
tokenTf = doc.get(token, 0)
score = tokenIdf * tokenTf
queryTfidf[tok] = score
return queryTfidf