def weightvectors(terms, documentvectors, progress=False):
n = len(documentvectors)
values = terms.values()
for t in values:
i = t.position
df = t.documentcount
for dv in documentvectors:
tf = dv[i]
dv[i] = int(round(term.tfidf(tf, n, df), 0))
if progress:
print ".",
sys.stdout.softspace = 0
def makevectors(documents, terms, progress=False):
documentvectors = []
nullvectors = []
for d in documents:
v = vec.Vec().fromlist(list(d.vector))
v.reference(d.id)
if v.norm() != 0:
values = terms.values()
for t in values:
tf = v[t.position]
df = t.documentcount
n = len(documents)
v[t.position] = int(round(term.tfidf(tf, n, df)))
documentvectors.append(v)
else:
nullvectors.append(v)
if progress:
print ".",
sys.stdout.softspace = 0
return documentvectors, nullvectors