def dump(self):
list = self._vector.items()
list = chew.sort(list, lambda x: x[1])
list.reverse()
for (term, count) in list:
print term.encode("utf-8"), count
print "Terms:", len(self._vector)
def dump(self):
list = self._vector.items()
list = chew.sort(list, lambda x: x[1])
list.reverse()
for (term, count) in list:
print term.encode("utf-8"), count
print "Terms:", len(self._vector)
def print_report(self):
print "TOTAL COUNT:", self._total
items = chew.sort(self._words.items(), lambda x: x[1])
items.reverse()
for (term, count) in items[:20]:
print "%30s %s" % (term.encode("utf-8"), count)
print "..."
for (term, count) in items[-20:]:
print "%30s %s" % (term.encode("utf-8"), count)
def print_report(self):
print "TOTAL COUNT:", self._total
items = chew.sort(self._words.items(), lambda x: x[1])
items.reverse()
for (term, count) in items[ : 20]:
print "%30s %s" % (term.encode("utf-8"), count)
print "..."
for (term, count) in items[-20 : ]:
print "%30s %s" % (term.encode("utf-8"), count)
def display_comparison(self, other, showall=0):
terms = {}
for (term, count) in self.get_pairs():
terms[term] = count * other.get_count(term)
termlist = chew.sort(terms.items(), lambda x: x[1])
termlist.reverse()
for (term, score) in termlist:
if score:
print(term, score)
if showall:
list = []
for (term, count) in self.get_pairs():
if not terms.get(term, 0):
list.append(term.encode("utf-8"))
print "+", string.join(list, ", ")
list = []
for (term, count) in other.get_pairs():
if not terms.get(term, 0):
list.append(term.encode("utf-8"))
print "-", string.join(list, ", ")
def display_comparison(self, other, showall = 0):
terms = {}
for (term, count) in self.get_pairs():
terms[term] = count * other.get_count(term)
termlist = chew.sort(terms.items(), lambda x: x[1])
termlist.reverse()
for (term, score) in termlist:
if score:
print (term, score)
if showall:
list = []
for (term, count) in self.get_pairs():
if not terms.get(term, 0):
list.append(term.encode("utf-8"))
print "+", string.join(list, ", ")
list = []
for (term, count) in other.get_pairs():
if not terms.get(term, 0):
list.append(term.encode("utf-8"))
print "-", string.join(list, ", ")
def k_nearest_neighbours_2(objects):
"""Makes clusters of objects. All objects must implement
object.compare(other), object.get_key(), and object.get_name()."""
pairs = []
for ix in range(len(objects)):
for i in range(ix+1, len(objects)):
pairs.append((objects[ix],
objects[i],
objects[ix].compare(objects[i])))
pairs = chew.sort(pairs, lambda x: x[2])
pairs.reverse()
clusters = []
clustermap = {}
for (t1, t2, score) in pairs:
if not score or (clustermap.has_key(t1.get_key()) and
clustermap.has_key(t2.get_key())):
if score:
pass #print "NOT USING:", (t1, t2, score)
continue
print (t1.get_name(), t2.get_name(), score)
#compare(get_matrix(t1, terms), get_matrix(t2, terms))
if clustermap.has_key(t1.get_key()):
c = clustermap[t1.get_key()]
elif clustermap.has_key(t2.get_key()):
c = clustermap[t2.get_key()]
else:
c = Cluster()
clusters.append(c)
for t in (t1, t2):
if not clustermap.has_key(t.get_key()):
c.add(t)
clustermap[t.get_key()] = c
return clusters
def k_nearest_neighbours_2(objects):
"""Makes clusters of objects. All objects must implement
object.compare(other), object.get_key(), and object.get_name()."""
pairs = []
for ix in range(len(objects)):
for i in range(ix + 1, len(objects)):
pairs.append(
(objects[ix], objects[i], objects[ix].compare(objects[i])))
pairs = chew.sort(pairs, lambda x: x[2])
pairs.reverse()
clusters = []
clustermap = {}
for (t1, t2, score) in pairs:
if not score or (clustermap.has_key(t1.get_key())
and clustermap.has_key(t2.get_key())):
if score:
pass #print "NOT USING:", (t1, t2, score)
continue
print(t1.get_name(), t2.get_name(), score)
#compare(get_matrix(t1, terms), get_matrix(t2, terms))
if clustermap.has_key(t1.get_key()):
c = clustermap[t1.get_key()]
elif clustermap.has_key(t2.get_key()):
c = clustermap[t2.get_key()]
else:
c = Cluster()
clusters.append(c)
for t in (t1, t2):
if not clustermap.has_key(t.get_key()):
c.add(t)
clustermap[t.get_key()] = c
return clusters