mirror of
https://github.com/explosion/spaCy.git
synced 2025-03-13 07:55:49 +03:00
Fix arc_eager oracle
This commit is contained in:
parent
a68d0e63f0
commit
420a986d15
|
@ -200,7 +200,6 @@ cdef class ArcEagerGold:
|
|||
sent_starts = example.get_aligned("SENT_START")
|
||||
assert len(heads) == len(labels) == len(sent_starts)
|
||||
self.c = create_gold_state(self.mem, stcls, heads, labels, sent_starts)
|
||||
self.update(stcls)
|
||||
|
||||
def update(self, StateClass stcls):
|
||||
update_gold_state(&self.c, stcls)
|
||||
|
@ -577,17 +576,12 @@ cdef class ArcEager(TransitionSystem):
|
|||
def is_gold_parse(self, StateClass state, gold):
|
||||
raise NotImplementedError
|
||||
|
||||
def has_gold(self, gold, start=0, end=None):
|
||||
raise NotImplementedError
|
||||
|
||||
def preprocess_gold(self, example):
|
||||
raise NotImplementedError
|
||||
|
||||
def init_gold_batch(self, examples):
|
||||
examples = [eg for eg in examples if self.has_gold(eg)]
|
||||
states = self.init_batch([eg.predicted for eg in examples])
|
||||
keeps = [i for i, s in enumerate(states) if not s.is_final()]
|
||||
states = [states[i] for i in keeps]
|
||||
golds = [ArcEagerGold(self, states[i], examples[i]) for i in keeps]
|
||||
states = [states[i] for i in keeps]
|
||||
for gold in golds:
|
||||
self._replace_unseen_labels(gold)
|
||||
n_steps = sum([len(s.queue) * 4 for s in states])
|
||||
|
@ -690,6 +684,9 @@ cdef class ArcEager(TransitionSystem):
|
|||
doc.is_parsed = True
|
||||
set_children_from_heads(doc.c, doc.length)
|
||||
|
||||
def has_gold(self, Example eg):
|
||||
return eg.y.is_parsed
|
||||
|
||||
cdef int set_valid(self, int* output, const StateC* st) nogil:
|
||||
cdef bint[N_MOVES] is_valid
|
||||
is_valid[SHIFT] = Shift.is_valid(st, 0)
|
||||
|
@ -736,21 +733,29 @@ cdef class ArcEager(TransitionSystem):
|
|||
raise ValueError
|
||||
|
||||
def get_oracle_sequence(self, Example example):
|
||||
cdef StateClass state
|
||||
cdef ArcEagerGold gold
|
||||
states, golds, n_steps = self.init_gold_batch([example])
|
||||
if not golds:
|
||||
return []
|
||||
|
||||
cdef Pool mem = Pool()
|
||||
# n_moves should not be zero at this point, but make sure to avoid zero-length mem alloc
|
||||
assert self.n_moves > 0
|
||||
costs = <float*>mem.alloc(self.n_moves, sizeof(float))
|
||||
is_valid = <int*>mem.alloc(self.n_moves, sizeof(int))
|
||||
|
||||
cdef StateClass state
|
||||
cdef ArcEagerGold gold
|
||||
states, golds, n_steps = self.init_gold_batch([example])
|
||||
state = states[0]
|
||||
gold = golds[0]
|
||||
history = []
|
||||
debug_log = []
|
||||
failed = False
|
||||
while not state.is_final():
|
||||
try:
|
||||
self.set_costs(is_valid, costs, state, gold)
|
||||
except ValueError:
|
||||
failed = True
|
||||
break
|
||||
for i in range(self.n_moves):
|
||||
if is_valid[i] and costs[i] <= 0:
|
||||
action = self.c[i]
|
||||
|
@ -766,6 +771,9 @@ cdef class ArcEager(TransitionSystem):
|
|||
action.do(state.c, action.label)
|
||||
break
|
||||
else:
|
||||
failed = False
|
||||
break
|
||||
if failed:
|
||||
print("Actions")
|
||||
for i in range(self.n_moves):
|
||||
print(self.get_class_name(i))
|
||||
|
|
Loading…
Reference in New Issue
Block a user