Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Allow multitask objectives to be added to the parser and NER more easily

This commit is contained in:
Matthew Honnibal 2018-01-21 19:37:02 +01:00
parent 4a7d524efb
commit 203d2ea830
2 changed files with 18 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
spacy/pipeline.pyx
View File

@ -882,14 +882,16 @@ cdef class DependencyParser(Parser):
def postprocesses(self): def postprocesses(self):
return [nonproj.deprojectivize] return [nonproj.deprojectivize]
def add_multitask_objective(self, target):
labeller = MultitaskObjective(self.vocab, target=target)
self._multitasks.append(labeller)
def init_multitask_objectives(self, gold_tuples, pipeline, sgd=None, **cfg): def init_multitask_objectives(self, gold_tuples, pipeline, sgd=None, **cfg):
for target in []: for labeller in self._multitasks:
labeller = MultitaskObjective(self.vocab, target=target)
tok2vec = self.model[0] tok2vec = self.model[0]
labeller.begin_training(gold_tuples, pipeline=pipeline, labeller.begin_training(gold_tuples, pipeline=pipeline,
tok2vec=tok2vec, sgd=sgd) tok2vec=tok2vec, sgd=sgd)
pipeline.append(labeller) pipeline.append((labeller.name, labeller))
self._multitasks.append(labeller)
def __reduce__(self): def __reduce__(self):
return (DependencyParser, (self.vocab, self.moves, self.model), return (DependencyParser, (self.vocab, self.moves, self.model),
@ -902,14 +904,16 @@ cdef class EntityRecognizer(Parser):
nr_feature = 6 nr_feature = 6
def add_multitask_objective(self, target):
labeller = MultitaskObjective(self.vocab, target=target)
self._multitasks.append(labeller)
def init_multitask_objectives(self, gold_tuples, pipeline, sgd=None, **cfg): def init_multitask_objectives(self, gold_tuples, pipeline, sgd=None, **cfg):
for target in []: for labeller in self._multitasks:
labeller = MultitaskObjective(self.vocab, target=target)
tok2vec = self.model[0] tok2vec = self.model[0]
labeller.begin_training(gold_tuples, pipeline=pipeline, labeller.begin_training(gold_tuples, pipeline=pipeline,
tok2vec=tok2vec) tok2vec=tok2vec)
pipeline.append(labeller) pipeline.append((labeller.name, labeller))
self._multitasks.append(labeller)
def __reduce__(self): def __reduce__(self):
return (EntityRecognizer, (self.vocab, self.moves, self.model), return (EntityRecognizer, (self.vocab, self.moves, self.model),

9
spacy/syntax/nn_parser.pyx
View File

@ -269,9 +269,6 @@ cdef class Parser:
zero_init(Affine(nr_class, hidden_width, drop_factor=0.0)) zero_init(Affine(nr_class, hidden_width, drop_factor=0.0))
) )
# TODO: This is an unfortunate hack atm!
# Used to set input dimensions in network.
lower.begin_training(lower.ops.allocate((500, token_vector_width)))
cfg = { cfg = {
'nr_class': nr_class, 'nr_class': nr_class,
'hidden_depth': depth, 'hidden_depth': depth,
@ -840,8 +837,14 @@ cdef class Parser:
self.cfg.update(cfg) self.cfg.update(cfg)
elif sgd is None: elif sgd is None:
sgd = self.create_optimizer() sgd = self.create_optimizer()
self.model[1].begin_training(
self.model[1].ops.allocate((5, cfg['token_vector_width'])))
return sgd return sgd
def add_multitask_objective(self, target):
# Defined in subclasses, to avoid circular import
raise NotImplementedError
def init_multitask_objectives(self, gold_tuples, pipeline, **cfg): def init_multitask_objectives(self, gold_tuples, pipeline, **cfg):
'''Setup models for secondary objectives, to benefit from multi-task '''Setup models for secondary objectives, to benefit from multi-task
learning. This method is intended to be overridden by subclasses. learning. This method is intended to be overridden by subclasses.