def tag_closure(tagdag):
"""Return data for the transitive closure of tag relationships.
The output is not in the same format as the input."""
result = {} # don't alter the original
bycat = util.group(tagdag, 'catID')
for catID, rels in bycat.iteritems(): #TODO(Py3k) use dict comprehension
result[catID] = KPA._close_one_cat(rels)
return result
def __iter__(self):
data_source = self.data_source
grouping = self.grouping
indices = list(range(len(data_source)))
if self.shuffle:
random.shuffle(indices)
group = {i: grouping(data_source[i]) for i in indices}
indices_pos = [i for i in indices if group[i]]
indices_neg = [i for i in indices if not group[i]]
pos_batch_size = self.pos_batch_size
neg_batch_size = self.neg_batch_size
batch_pos = util.group(pos_batch_size, indices_pos)
batch_neg = util.group(neg_batch_size, indices_neg)
for x, y in zip(batch_pos, batch_neg):
yield from x
yield from y
def test_group(self):
"""Grouping a list of dicts by one key's value."""
# empty
self.assert_eq_rec({}, util.group([], "hello"))
#bad key
self.expect_error(lambda:util.group([{}], "imaginary"), KeyError)
#normal
start = [
{0:0,1:0},{0:0,1:5},{0:0,1:2},{0:4,1:2},{0:3,1:2},{0:8,1:9},
{0:3,1:3},{0:0,1:7},{0:4,1:6},{0:4,1:7},{0:0,1:8},{0:8,1:9}
]
expected = {
0:[{0:0,1:0},{0:0,1:5},{0:0,1:2},{0:0,1:7},{0:0,1:8}],
4:[{0:4,1:2},{0:4,1:6},{0:4,1:7}],
3:[{0:3,1:2},{0:3,1:3}],
8:[{0:8,1:9},{0:8,1:9}]
}
actual = util.group(start, 0)
self.assert_eq_rec(expected, actual)
