def fromstring(cls, lex_str, ontology=None, include_semantics=False,
default_weight=0.001):
"""
Convert string representation into a lexicon for CCGs.
"""
ccg_lexicon.CCGVar.reset_id()
primitives, starts = [], []
families = {}
entries = defaultdict(list)
for line in lex_str.splitlines():
# Strip comments and leading/trailing whitespace.
line = COMMENTS_RE.match(line).groups()[0].strip()
if line == "":
continue
if line.startswith(':-'):
# A line of primitive categories.
# The first one is the target category
# ie, :- S, N, NP, VP
primitives = primitives + [prim.strip() for prim in line[2:].strip().split(',')]
# But allow multiple target categories separated by a colon in the first element:
# ie, :- S:N,NP,VP
starts = primitives[0].split(":")
primitives = starts + primitives[1:]
else:
# Either a family definition, or a word definition
(ident, sep, rhs) = LEX_RE.match(line).groups()
(catstr, semantics_str, weight) = RHS_RE.match(rhs).groups()
(cat, var) = ccg_lexicon.augParseCategory(catstr, primitives, families)
if sep == '::':
# Family definition
# ie, Det :: NP/N
families[ident] = (cat, var)
# TODO weight?
else:
semantics = None
if include_semantics is True:
if semantics_str is None:
raise AssertionError(line + " must contain semantics because include_semantics is set to True")
else:
semantics = l.Expression.fromstring(ccg_lexicon.SEMANTICS_RE.match(semantics_str).groups()[0])
# Assign types.
if ontology is not None:
ontology.typecheck(semantics)
weight = float(weight[1:-1]) if weight is not None else default_weight
weight = Parameter(T.tensor(weight, requires_grad=True))
# Word definition
# ie, which => (N\N)/(S/NP)
entries[ident].append(Token(ident, cat, semantics, weight=weight))
return cls(starts, primitives, families, entries, has_semantics=include_semantics,
ontology=ontology)
def parse_category(self, cat_str):
return ccg_lexicon.augParseCategory(cat_str, self._primitives,
self._families)[0]
def test_case(cat, expected):
eq_(get_semantic_arity(
augParseCategory(cat, lex._primitives, lex._families)[0]),
expected,
msg=str(cat))