def __iter__(self):
it = iter(self.rawIter())
(history, predicted), value = it.next()
values = {predicted: value}
for (h, p), v in it:
if h == history:
values[p] = values.get(p, 0) + v
elif h > history:
for predicted, value in sorted(values.iteritems()):
yield (history, predicted), value
history = h
values = {p: v}
else:
raise ValueError('sequence not ordered', history, h)
for predicted, value in sorted(values.iteritems()):
yield (history, predicted), value
def __iter__(self):
it = iter(self.rawIter())
(history, predicted), value = it.next()
values = { predicted: value }
for (h, p), v in it:
if h == history:
values[p] = values.get(p, 0) + v
elif h > history:
for predicted, value in sorted(values.iteritems()):
yield (history, predicted), value
history = h
values = { p: v }
else:
raise ValueError('sequence not ordered', history, h)
for predicted, value in sorted(values.iteritems()):
yield (history, predicted), value
def get_free_vars(self):
if not self.nested:
return ()
free = self.frees | (self.uses - (self.globals | set(self.defs)))
if not free:
return ()
return sorted(list(free))
def testIter(self):
sv = sparse(self.example)
ref = sorted(self.example)
for i, item in enumerate(sv):
self.failUnlessEqual(item, ref[i])
sl = list(sv)
self.failUnlessEqual(sl, ref)
def testIter(self):
sv = sparse(self.example)
ref = sorted(self.example)
for i, item in enumerate(sv):
self.failUnlessEqual(item, ref[i])
sl = list(sv)
self.failUnlessEqual(sl, ref)
def countsOfCounts(counts, group = None):
histogram = {}
counts = assertIsConsolidated(counts)
for gram, count in counts:
cat = group and group(count) or count
try:
histogram[cat] += 1
except KeyError:
histogram[cat] = 1
result = sorted(histogram.items())
return result
def countsOfCounts(counts, group=None):
histogram = {}
counts = assertIsConsolidated(counts)
for gram, count in counts:
cat = group and group(count) or count
try:
histogram[cat] += 1
except KeyError:
histogram[cat] = 1
result = sorted(histogram.items())
return result
def sizeTemplates(self):
result = set()
for i in range(1, self.size() + 1):
left, right = self.symbol(i)
result.add((len(left), len(right)))
return sorted(result)
def dictAdd(self, a, b):
result = dict(a)
for k, v in b:
result[k] = result.get(k, 0) + v
return sorted(result.items())
def testThreshold(self):
for t in [0.5, 1.0, 1.5, 2.0, 2.5]:
sv = sparse(self.example).threshold(t)
self.failUnlessEqual(list(sv), sorted(filter(lambda x: x[1]>=t, self.example)))
def joined(m):
if m + 1 < M:
return outerJoin(sorted(probabilities(m)),
sorted(backOffs(m + 1)))
else:
return outerJoin(sorted(probabilities(m)), [])
def sizeTemplates(self):
result = set()
for i in range(1, self.size() + 1):
left, right = self.symbol(i)
result.add((len(left), len(right)))
return sorted(result)
def dictAdd(self, a, b):
result = dict(a)
for k, v in b:
result[k] = result.get(k, 0) + v
return sorted(result.items())
def testThreshold(self):
for t in [0.5, 1.0, 1.5, 2.0, 2.5]:
sv = sparse(self.example).threshold(t)
self.failUnlessEqual(
list(sv), sorted(filter(lambda x: x[1] >= t, self.example)))
def testThreshold(self):
for t in [0.5, 1.0, 1.5, 2.0, 2.5]:
sv = sparse(self.example).threshold(t)
self.assertEqual(list(sv),
sorted([x for x in self.example if x[1] >= t]))
def joined(m): if m+1 < M: return outerJoin(sorted(probabilities(m)), sorted(backOffs(m+1))) else: return outerJoin(sorted(probabilities(m)), [])