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

💫 Break up large pipeline.pyx (#3246) 

* Break up large pipeline.pyx

* Merge some components back together

* Fix typo
This commit is contained in:
Ines Montani 2019-02-10 12:14:51 +01:00 committed by GitHub
parent e7593b791e
commit a9f8d17632
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
7 changed files with 400 additions and 360 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
setup.py
View File

@ -41,7 +41,7 @@ MOD_NAMES = [
"spacy.vocab",
"spacy.attrs",
"spacy.morphology",
"spacy.pipeline",
"spacy.pipeline.pipes",
"spacy.syntax.stateclass",
"spacy.syntax._state",
"spacy.tokenizer",

0
spacy/pipeline.pxd
View File

8
spacy/pipeline/__init__.py Normal file
View File

@ -0,0 +1,8 @@
# coding: utf8
from __future__ import unicode_literals
from .pipes import Tagger, DependencyParser, EntityRecognizer # noqa
from .pipes import TextCategorizer, Tensorizer, Pipe # noqa
from .entityruler import EntityRuler # noqa
from .hooks import SentenceSegmenter, SimilarityHook # noqa
from .functions import merge_entities, merge_noun_chunks, merge_subtokens # noqa

170
spacy/pipeline/entityruler.py Normal file
View File

@ -0,0 +1,170 @@
# coding: utf8
from __future__ import unicode_literals
from collections import defaultdict
import srsly
from ..errors import Errors
from ..compat import basestring_
from ..util import ensure_path
from ..tokens import Span
from ..matcher import Matcher, PhraseMatcher
class EntityRuler(object):
name = "entity_ruler"
def __init__(self, nlp, **cfg):
"""Initialise the entitiy ruler. If patterns are supplied here, they
need to be a list of dictionaries with a `"label"` and `"pattern"`
key. A pattern can either be a token pattern (list) or a phrase pattern
(string). For example: `{'label': 'ORG', 'pattern': 'Apple'}`.
nlp (Language): The shared nlp object to pass the vocab to the matchers
and process phrase patterns.
patterns (iterable): Optional patterns to load in.
overwrite_ents (bool): If existing entities are present, e.g. entities
added by the model, overwrite them by matches if necessary.
**cfg: Other config parameters. If pipeline component is loaded as part
of a model pipeline, this will include all keyword arguments passed
to `spacy.load`.
RETURNS (EntityRuler): The newly constructed object.
"""
self.nlp = nlp
self.overwrite = cfg.get("overwrite_ents", False)
self.token_patterns = defaultdict(list)
self.phrase_patterns = defaultdict(list)
self.matcher = Matcher(nlp.vocab)
self.phrase_matcher = PhraseMatcher(nlp.vocab)
patterns = cfg.get("patterns")
if patterns is not None:
self.add_patterns(patterns)
def __len__(self):
"""The number of all patterns added to the entity ruler."""
n_token_patterns = sum(len(p) for p in self.token_patterns.values())
n_phrase_patterns = sum(len(p) for p in self.phrase_patterns.values())
return n_token_patterns + n_phrase_patterns
def __contains__(self, label):
"""Whether a label is present in the patterns."""
return label in self.token_patterns or label in self.phrase_patterns
def __call__(self, doc):
"""Find matches in document and add them as entities.
doc (Doc): The Doc object in the pipeline.
RETURNS (Doc): The Doc with added entities, if available.
"""
matches = list(self.matcher(doc)) + list(self.phrase_matcher(doc))
matches = set(
[(m_id, start, end) for m_id, start, end in matches if start != end]
)
get_sort_key = lambda m: (m[2] - m[1], m[1])
matches = sorted(matches, key=get_sort_key, reverse=True)
entities = list(doc.ents)
new_entities = []
seen_tokens = set()
for match_id, start, end in matches:
if any(t.ent_type for t in doc[start:end]) and not self.overwrite:
continue
# check for end - 1 here because boundaries are inclusive
if start not in seen_tokens and end - 1 not in seen_tokens:
new_entities.append(Span(doc, start, end, label=match_id))
entities = [
e for e in entities if not (e.start < end and e.end > start)
]
seen_tokens.update(range(start, end))
doc.ents = entities + new_entities
return doc
@property
def labels(self):
"""All labels present in the match patterns.
RETURNS (set): The string labels.
"""
all_labels = set(self.token_patterns.keys())
all_labels.update(self.phrase_patterns.keys())
return all_labels
@property
def patterns(self):
"""Get all patterns that were added to the entity ruler.
RETURNS (list): The original patterns, one dictionary per pattern.
"""
all_patterns = []
for label, patterns in self.token_patterns.items():
for pattern in patterns:
all_patterns.append({"label": label, "pattern": pattern})
for label, patterns in self.phrase_patterns.items():
for pattern in patterns:
all_patterns.append({"label": label, "pattern": pattern.text})
return all_patterns
def add_patterns(self, patterns):
"""Add patterns to the entitiy ruler. A pattern can either be a token
pattern (list of dicts) or a phrase pattern (string). For example:
{'label': 'ORG', 'pattern': 'Apple'}
{'label': 'GPE', 'pattern': [{'lower': 'san'}, {'lower': 'francisco'}]}
patterns (list): The patterns to add.
"""
for entry in patterns:
label = entry["label"]
pattern = entry["pattern"]
if isinstance(pattern, basestring_):
self.phrase_patterns[label].append(self.nlp(pattern))
elif isinstance(pattern, list):
self.token_patterns[label].append(pattern)
else:
raise ValueError(Errors.E097.format(pattern=pattern))
for label, patterns in self.token_patterns.items():
self.matcher.add(label, None, *patterns)
for label, patterns in self.phrase_patterns.items():
self.phrase_matcher.add(label, None, *patterns)
def from_bytes(self, patterns_bytes, **kwargs):
"""Load the entity ruler from a bytestring.
patterns_bytes (bytes): The bytestring to load.
**kwargs: Other config paramters, mostly for consistency.
RETURNS (EntityRuler): The loaded entity ruler.
"""
patterns = srsly.msgpack_loads(patterns_bytes)
self.add_patterns(patterns)
return self
def to_bytes(self, **kwargs):
"""Serialize the entity ruler patterns to a bytestring.
RETURNS (bytes): The serialized patterns.
"""
return srsly.msgpack_dumps(self.patterns)
def from_disk(self, path, **kwargs):
"""Load the entity ruler from a file. Expects a file containing
newline-delimited JSON (JSONL) with one entry per line.
path (unicode / Path): The JSONL file to load.
**kwargs: Other config paramters, mostly for consistency.
RETURNS (EntityRuler): The loaded entity ruler.
"""
path = ensure_path(path)
path = path.with_suffix(".jsonl")
patterns = srsly.read_jsonl(path)
self.add_patterns(patterns)
return self
def to_disk(self, path, **kwargs):
"""Save the entity ruler patterns to a directory. The patterns will be
saved as newline-delimited JSON (JSONL).
path (unicode / Path): The JSONL file to load.
**kwargs: Other config paramters, mostly for consistency.
RETURNS (EntityRuler): The loaded entity ruler.
"""
path = ensure_path(path)
path = path.with_suffix(".jsonl")
srsly.write_jsonl(path, self.patterns)

48
spacy/pipeline/functions.py Normal file
View File

@ -0,0 +1,48 @@
# coding: utf8
from __future__ import unicode_literals
from ..matcher import Matcher
# TODO: replace doc.merge with doc.retokenize
def merge_noun_chunks(doc):
"""Merge noun chunks into a single token.
doc (Doc): The Doc object.
RETURNS (Doc): The Doc object with merged noun chunks.
"""
if not doc.is_parsed:
return doc
spans = [
(np.start_char, np.end_char, np.root.tag, np.root.dep) for np in doc.noun_chunks
]
for start, end, tag, dep in spans:
doc.merge(start, end, tag=tag, dep=dep)
return doc
def merge_entities(doc):
"""Merge entities into a single token.
doc (Doc): The Doc object.
RETURNS (Doc): The Doc object with merged noun entities.
"""
spans = [
(e.start_char, e.end_char, e.root.tag, e.root.dep, e.label) for e in doc.ents
]
for start, end, tag, dep, ent_type in spans:
doc.merge(start, end, tag=tag, dep=dep, ent_type=ent_type)
return doc
def merge_subtokens(doc, label="subtok"):
merger = Matcher(doc.vocab)
merger.add("SUBTOK", None, [{"DEP": label, "op": "+"}])
matches = merger(doc)
spans = [doc[start : end + 1] for _, start, end in matches]
offsets = [(span.start_char, span.end_char) for span in spans]
for start_char, end_char in offsets:
doc.merge(start_char, end_char)
return doc

100
spacy/pipeline/hooks.py Normal file
View File

@ -0,0 +1,100 @@
# coding: utf8
from __future__ import unicode_literals
from thinc.t2v import Pooling, max_pool, mean_pool
from thinc.neural._classes.difference import Siamese, CauchySimilarity
from .pipes import Pipe
from .._ml import link_vectors_to_models
class SentenceSegmenter(object):
"""A simple spaCy hook, to allow custom sentence boundary detection logic
(that doesn't require the dependency parse). To change the sentence
boundary detection strategy, pass a generator function `strategy` on
initialization, or assign a new strategy to the .strategy attribute.
Sentence detection strategies should be generators that take `Doc` objects
and yield `Span` objects for each sentence.
"""
name = "sentencizer"
def __init__(self, vocab, strategy=None):
self.vocab = vocab
if strategy is None or strategy == "on_punct":
strategy = self.split_on_punct
self.strategy = strategy
def __call__(self, doc):
doc.user_hooks["sents"] = self.strategy
return doc
@staticmethod
def split_on_punct(doc):
start = 0
seen_period = False
for i, word in enumerate(doc):
if seen_period and not word.is_punct:
yield doc[start : word.i]
start = word.i
seen_period = False
elif word.text in [".", "!", "?"]:
seen_period = True
if start < len(doc):
yield doc[start : len(doc)]
class SimilarityHook(Pipe):
"""
Experimental: A pipeline component to install a hook for supervised
similarity into `Doc` objects. Requires a `Tensorizer` to pre-process
documents. The similarity model can be any object obeying the Thinc `Model`
interface. By default, the model concatenates the elementwise mean and
elementwise max of the two tensors, and compares them using the
Cauchy-like similarity function from Chen (2013):
>>> similarity = 1. / (1. + (W * (vec1-vec2)**2).sum())
Where W is a vector of dimension weights, initialized to 1.
"""
name = "similarity"
def __init__(self, vocab, model=True, **cfg):
self.vocab = vocab
self.model = model
self.cfg = dict(cfg)
@classmethod
def Model(cls, length):
return Siamese(Pooling(max_pool, mean_pool), CauchySimilarity(length))
def __call__(self, doc):
"""Install similarity hook"""
doc.user_hooks["similarity"] = self.predict
return doc
def pipe(self, docs, **kwargs):
for doc in docs:
yield self(doc)
def predict(self, doc1, doc2):
self.require_model()
return self.model.predict([(doc1, doc2)])
def update(self, doc1_doc2, golds, sgd=None, drop=0.0):
self.require_model()
sims, bp_sims = self.model.begin_update(doc1_doc2, drop=drop)
def begin_training(self, _=tuple(), pipeline=None, sgd=None, **kwargs):
"""Allocate model, using width from tensorizer in pipeline.
gold_tuples (iterable): Gold-standard training data.
pipeline (list): The pipeline the model is part of.
"""
if self.model is True:
self.model = self.Model(pipeline[0].model.nO)
link_vectors_to_models(self.vocab)
if sgd is None:
sgd = self.create_optimizer()
return sgd

432
spacy/pipeline.pyx → spacy/pipeline/pipes.pyx
View File

@ -3,271 +3,39 @@
# coding: utf8
from __future__ import unicode_literals
import numpy
cimport numpy as np
from collections import OrderedDict, defaultdict
import numpy
from collections import OrderedDict
import srsly
from thinc.api import chain
from thinc.v2v import Affine, Maxout, Softmax
from thinc.misc import LayerNorm
from thinc.t2v import Pooling, max_pool, mean_pool
from thinc.neural.util import to_categorical, copy_array
from thinc.neural._classes.difference import Siamese, CauchySimilarity
from .tokens.doc cimport Doc
from .syntax.nn_parser cimport Parser
from .syntax import nonproj
from .syntax.ner cimport BiluoPushDown
from .syntax.arc_eager cimport ArcEager
from .morphology cimport Morphology
from .vocab cimport Vocab
from .syntax import nonproj
from .matcher import Matcher
from ..tokens.doc cimport Doc
from ..syntax.nn_parser cimport Parser
from ..syntax.ner cimport BiluoPushDown
from ..syntax.arc_eager cimport ArcEager
from ..morphology cimport Morphology
from ..vocab cimport Vocab
from .matcher import Matcher, PhraseMatcher
from .tokens.span import Span
from .attrs import POS, ID
from .parts_of_speech import X
from ._ml import Tok2Vec, build_text_classifier, build_tagger_model
from ._ml import build_simple_cnn_text_classifier
from ._ml import link_vectors_to_models, zero_init, flatten
from ._ml import create_default_optimizer
from ._ml import masked_language_model
from .errors import Errors, TempErrors
from .compat import basestring_
from . import util
from ..syntax import nonproj
from ..attrs import POS, ID
from ..parts_of_speech import X
from .._ml import Tok2Vec, build_tagger_model, build_simple_cnn_text_classifier
from .._ml import link_vectors_to_models, zero_init, flatten
from .._ml import masked_language_model, create_default_optimizer
from ..errors import Errors, TempErrors
from .. import util
class SentenceSegmenter(object):
"""A simple spaCy hook, to allow custom sentence boundary detection logic
(that doesn't require the dependency parse). To change the sentence
boundary detection strategy, pass a generator function `strategy` on
initialization, or assign a new strategy to the .strategy attribute.
Sentence detection strategies should be generators that take `Doc` objects
and yield `Span` objects for each sentence.
"""
name = 'sentencizer'
def __init__(self, vocab, strategy=None):
self.vocab = vocab
if strategy is None or strategy == 'on_punct':
strategy = self.split_on_punct
self.strategy = strategy
def __call__(self, doc):
doc.user_hooks['sents'] = self.strategy
return doc
@staticmethod
def split_on_punct(doc):
start = 0
seen_period = False
for i, word in enumerate(doc):
if seen_period and not word.is_punct:
yield doc[start:word.i]
start = word.i
seen_period = False
elif word.text in ['.', '!', '?']:
seen_period = True
if start < len(doc):
yield doc[start:len(doc)]
def merge_noun_chunks(doc):
"""Merge noun chunks into a single token.
doc (Doc): The Doc object.
RETURNS (Doc): The Doc object with merged noun chunks.
"""
if not doc.is_parsed:
return doc
spans = [(np.start_char, np.end_char, np.root.tag, np.root.dep)
for np in doc.noun_chunks]
for start, end, tag, dep in spans:
doc.merge(start, end, tag=tag, dep=dep)
return doc
def merge_entities(doc):
"""Merge entities into a single token.
doc (Doc): The Doc object.
RETURNS (Doc): The Doc object with merged noun entities.
"""
spans = [(e.start_char, e.end_char, e.root.tag, e.root.dep, e.label)
for e in doc.ents]
for start, end, tag, dep, ent_type in spans:
doc.merge(start, end, tag=tag, dep=dep, ent_type=ent_type)
return doc
def merge_subtokens(doc, label='subtok'):
merger = Matcher(doc.vocab)
merger.add('SUBTOK', None, [{'DEP': label, 'op': '+'}])
matches = merger(doc)
spans = [doc[start:end+1] for _, start, end in matches]
offsets = [(span.start_char, span.end_char) for span in spans]
for start_char, end_char in offsets:
doc.merge(start_char, end_char)
return doc
class EntityRuler(object):
name = 'entity_ruler'
def __init__(self, nlp, **cfg):
"""Initialise the entitiy ruler. If patterns are supplied here, they
need to be a list of dictionaries with a `"label"` and `"pattern"`
key. A pattern can either be a token pattern (list) or a phrase pattern
(string). For example: `{'label': 'ORG', 'pattern': 'Apple'}`.
nlp (Language): The shared nlp object to pass the vocab to the matchers
and process phrase patterns.
patterns (iterable): Optional patterns to load in.
overwrite_ents (bool): If existing entities are present, e.g. entities
added by the model, overwrite them by matches if necessary.
**cfg: Other config parameters. If pipeline component is loaded as part
of a model pipeline, this will include all keyword arguments passed
to `spacy.load`.
RETURNS (EntityRuler): The newly constructed object.
"""
self.nlp = nlp
self.overwrite = cfg.get('overwrite_ents', False)
self.token_patterns = defaultdict(list)
self.phrase_patterns = defaultdict(list)
self.matcher = Matcher(nlp.vocab)
self.phrase_matcher = PhraseMatcher(nlp.vocab)
patterns = cfg.get('patterns')
if patterns is not None:
self.add_patterns(patterns)
def __len__(self):
"""The number of all patterns added to the entity ruler."""
n_token_patterns = sum(len(p) for p in self.token_patterns.values())
n_phrase_patterns = sum(len(p) for p in self.phrase_patterns.values())
return n_token_patterns + n_phrase_patterns
def __contains__(self, label):
"""Whether a label is present in the patterns."""
return label in self.token_patterns or label in self.phrase_patterns
def __call__(self, doc):
"""Find matches in document and add them as entities.
doc (Doc): The Doc object in the pipeline.
RETURNS (Doc): The Doc with added entities, if available.
"""
matches = list(self.matcher(doc)) + list(self.phrase_matcher(doc))
matches = set([(m_id, start, end) for m_id, start, end in matches
if start != end])
get_sort_key = lambda m: (m[2] - m[1], m[1])
matches = sorted(matches, key=get_sort_key, reverse=True)
entities = list(doc.ents)
new_entities = []
seen_tokens = set()
for match_id, start, end in matches:
if any(t.ent_type for t in doc[start:end]) and not self.overwrite:
continue
# check for end - 1 here because boundaries are inclusive
if start not in seen_tokens and end - 1 not in seen_tokens:
new_entities.append(Span(doc, start, end, label=match_id))
entities = [e for e in entities
if not (e.start < end and e.end > start)]
seen_tokens.update(range(start, end))
doc.ents = entities + new_entities
return doc
@property
def labels(self):
"""All labels present in the match patterns.
RETURNS (set): The string labels.
"""
all_labels = set(self.token_patterns.keys())
all_labels.update(self.phrase_patterns.keys())
return all_labels
@property
def patterns(self):
"""Get all patterns that were added to the entity ruler.
RETURNS (list): The original patterns, one dictionary per pattern.
"""
all_patterns = []
for label, patterns in self.token_patterns.items():
for pattern in patterns:
all_patterns.append({'label': label, 'pattern': pattern})
for label, patterns in self.phrase_patterns.items():
for pattern in patterns:
all_patterns.append({'label': label, 'pattern': pattern.text})
return all_patterns
def add_patterns(self, patterns):
"""Add patterns to the entitiy ruler. A pattern can either be a token
pattern (list of dicts) or a phrase pattern (string). For example:
{'label': 'ORG', 'pattern': 'Apple'}
{'label': 'GPE', 'pattern': [{'lower': 'san'}, {'lower': 'francisco'}]}
patterns (list): The patterns to add.
"""
for entry in patterns:
label = entry['label']
pattern = entry['pattern']
if isinstance(pattern, basestring_):
self.phrase_patterns[label].append(self.nlp(pattern))
elif isinstance(pattern, list):
self.token_patterns[label].append(pattern)
else:
raise ValueError(Errors.E097.format(pattern=pattern))
for label, patterns in self.token_patterns.items():
self.matcher.add(label, None, *patterns)
for label, patterns in self.phrase_patterns.items():
self.phrase_matcher.add(label, None, *patterns)
def from_bytes(self, patterns_bytes, **kwargs):
"""Load the entity ruler from a bytestring.
patterns_bytes (bytes): The bytestring to load.
**kwargs: Other config paramters, mostly for consistency.
RETURNS (EntityRuler): The loaded entity ruler.
"""
patterns = srsly.msgpack_loads(patterns_bytes)
self.add_patterns(patterns)
return self
def to_bytes(self, **kwargs):
"""Serialize the entity ruler patterns to a bytestring.
RETURNS (bytes): The serialized patterns.
"""
return srsly.msgpack_dumps(self.patterns)
def from_disk(self, path, **kwargs):
"""Load the entity ruler from a file. Expects a file containing
newline-delimited JSON (JSONL) with one entry per line.
path (unicode / Path): The JSONL file to load.
**kwargs: Other config paramters, mostly for consistency.
RETURNS (EntityRuler): The loaded entity ruler.
"""
path = util.ensure_path(path)
path = path.with_suffix('.jsonl')
patterns = srsly.read_jsonl(path)
self.add_patterns(patterns)
return self
def to_disk(self, path, **kwargs):
"""Save the entity ruler patterns to a directory. The patterns will be
saved as newline-delimited JSON (JSONL).
path (unicode / Path): The JSONL file to load.
**kwargs: Other config paramters, mostly for consistency.
RETURNS (EntityRuler): The loaded entity ruler.
"""
path = util.ensure_path(path)
path = path.with_suffix('.jsonl')
srsly.write_jsonl(path, self.patterns)
def _load_cfg(path):
if path.exists():
return srsly.read_json(path)
else:
return {}
class Pipe(object):
@ -275,6 +43,7 @@ class Pipe(object):
it defines the interface that components should follow to function as
components in a spaCy analysis pipeline.
"""
name = None
@classmethod
@ -300,7 +69,7 @@ class Pipe(object):
def require_model(self):
"""Raise an error if the component's model is not initialized."""
if getattr(self, 'model', None) in (None, True, False):
if getattr(self, "model", None) in (None, True, False):
raise ValueError(Errors.E109.format(name=self.name))
def pipe(self, stream, batch_size=128, n_threads=-1):
@ -326,7 +95,7 @@ class Pipe(object):
"""Modify a batch of documents, using pre-computed scores."""
raise NotImplementedError
def update(self, docs, golds, drop=0., sgd=None, losses=None):
def update(self, docs, golds, drop=0.0, sgd=None, losses=None):
"""Learn from a batch of documents and gold-standard information,
updating the pipe's model.
@ -353,11 +122,11 @@ class Pipe(object):
raise NotImplementedError
def create_optimizer(self):
return create_default_optimizer(self.model.ops,
**self.cfg.get('optimizer', {}))
return create_default_optimizer(self.model.ops, **self.cfg.get("optimizer", {}))
def begin_training(self, get_gold_tuples=lambda: [], pipeline=None, sgd=None,
**kwargs):
def begin_training(
self, get_gold_tuples=lambda: [], pipeline=None, sgd=None, **kwargs
):
"""Initialize the pipe for training, using data exampes if available.
If no model has been initialized yet, the model is added."""
if self.model is True:
@ -375,70 +144,70 @@ class Pipe(object):
def to_bytes(self, **exclude):
"""Serialize the pipe to a bytestring."""
serialize = OrderedDict()
serialize['cfg'] = lambda: srsly.json_dumps(self.cfg)
serialize["cfg"] = lambda: srsly.json_dumps(self.cfg)
if self.model in (True, False, None):
serialize['model'] = lambda: self.model
serialize["model"] = lambda: self.model
else:
serialize['model'] = self.model.to_bytes
serialize['vocab'] = self.vocab.to_bytes
serialize["model"] = self.model.to_bytes
serialize["vocab"] = self.vocab.to_bytes
return util.to_bytes(serialize, exclude)
def from_bytes(self, bytes_data, **exclude):
"""Load the pipe from a bytestring."""
def load_model(b):
# TODO: Remove this once we don't have to handle previous models
if self.cfg.get('pretrained_dims') and 'pretrained_vectors' not in self.cfg:
self.cfg['pretrained_vectors'] = self.vocab.vectors.name
if self.cfg.get("pretrained_dims") and "pretrained_vectors" not in self.cfg:
self.cfg["pretrained_vectors"] = self.vocab.vectors.name
if self.model is True:
self.model = self.Model(**self.cfg)
self.model.from_bytes(b)
deserialize = OrderedDict((
('cfg', lambda b: self.cfg.update(srsly.json_loads(b))),
('vocab', lambda b: self.vocab.from_bytes(b)),
('model', load_model),
))
deserialize = OrderedDict(
(
("cfg", lambda b: self.cfg.update(srsly.json_loads(b))),
("vocab", lambda b: self.vocab.from_bytes(b)),
("model", load_model),
)
)
util.from_bytes(bytes_data, deserialize, exclude)
return self
def to_disk(self, path, **exclude):
"""Serialize the pipe to disk."""
serialize = OrderedDict()
serialize['cfg'] = lambda p: srsly.write_json(p, self.cfg)
serialize['vocab'] = lambda p: self.vocab.to_disk(p)
serialize["cfg"] = lambda p: srsly.write_json(p, self.cfg)
serialize["vocab"] = lambda p: self.vocab.to_disk(p)
if self.model not in (None, True, False):
serialize['model'] = lambda p: p.open('wb').write(self.model.to_bytes())
serialize["model"] = lambda p: p.open("wb").write(self.model.to_bytes())
util.to_disk(path, serialize, exclude)
def from_disk(self, path, **exclude):
"""Load the pipe from disk."""
def load_model(p):
# TODO: Remove this once we don't have to handle previous models
if self.cfg.get('pretrained_dims') and 'pretrained_vectors' not in self.cfg:
self.cfg['pretrained_vectors'] = self.vocab.vectors.name
if self.cfg.get("pretrained_dims") and "pretrained_vectors" not in self.cfg:
self.cfg["pretrained_vectors"] = self.vocab.vectors.name
if self.model is True:
self.model = self.Model(**self.cfg)
self.model.from_bytes(p.open('rb').read())
self.model.from_bytes(p.open("rb").read())
deserialize = OrderedDict((
('cfg', lambda p: self.cfg.update(_load_cfg(p))),
('vocab', lambda p: self.vocab.from_disk(p)),
('model', load_model),
))
deserialize = OrderedDict(
(
("cfg", lambda p: self.cfg.update(_load_cfg(p))),
("vocab", lambda p: self.vocab.from_disk(p)),
("model", load_model),
)
)
util.from_disk(path, deserialize, exclude)
return self
def _load_cfg(path):
if path.exists():
return srsly.read_json(path)
else:
return {}
class Tensorizer(Pipe):
"""Pre-train position-sensitive vectors for tokens."""
name = 'tensorizer'
name = "tensorizer"
@classmethod
def Model(cls, output_size=300, **cfg):
@ -449,7 +218,7 @@ class Tensorizer(Pipe):
**cfg: Config parameters.
RETURNS (Model): A `thinc.neural.Model` or similar instance.
"""
input_size = util.env_opt('token_vector_width', cfg.get('input_size', 96))
input_size = util.env_opt("token_vector_width", cfg.get("input_size", 96))
return zero_init(Affine(output_size, input_size, drop_factor=0.0))
def __init__(self, vocab, model=True, **cfg):
@ -470,7 +239,7 @@ class Tensorizer(Pipe):
self.model = model
self.input_models = []
self.cfg = dict(cfg)
self.cfg.setdefault('cnn_maxout_pieces', 3)
self.cfg.setdefault("cnn_maxout_pieces", 3)
def __call__(self, doc):
"""Add context-sensitive vectors to a `Doc`, e.g. from a CNN or LSTM
@ -516,10 +285,12 @@ class Tensorizer(Pipe):
"""
for doc, tensor in zip(docs, tensors):
if tensor.shape[0] != len(doc):
raise ValueError(Errors.E076.format(rows=tensor.shape[0], words=len(doc)))
raise ValueError(
Errors.E076.format(rows=tensor.shape[0], words=len(doc))
)
doc.tensor = tensor
def update(self, docs, golds, state=None, drop=0., sgd=None, losses=None):
def update(self, docs, golds, state=None, drop=0.0, sgd=None, losses=None):
"""Update the model.
docs (iterable): A batch of `Doc` objects.
@ -545,7 +316,7 @@ class Tensorizer(Pipe):
for d_input, bp_input in zip(d_inputs, bp_inputs):
bp_input(d_input, sgd=sgd)
if losses is not None:
losses.setdefault(self.name, 0.)
losses.setdefault(self.name, 0.0)
losses[self.name] += loss
return loss
@ -553,11 +324,10 @@ class Tensorizer(Pipe):
ids = self.model.ops.flatten([doc.to_array(ID).ravel() for doc in docs])
target = self.vocab.vectors.data[ids]
d_scores = (prediction - target) / prediction.shape[0]
loss = (d_scores**2).sum()
loss = (d_scores ** 2).sum()
return loss, d_scores
def begin_training(self, gold_tuples=lambda: [], pipeline=None, sgd=None,
**kwargs):
def begin_training(self, gold_tuples=lambda: [], pipeline=None, sgd=None, **kwargs):
"""Allocate models, pre-process training data and acquire an
optimizer.
@ -566,7 +336,7 @@ class Tensorizer(Pipe):
"""
if pipeline is not None:
for name, model in pipeline:
if getattr(model, 'tok2vec', None):
if getattr(model, "tok2vec", None):
self.input_models.append(model.tok2vec)
if self.model is True:
self.model = self.Model(**self.cfg)
@ -1086,61 +856,6 @@ class ClozeMultitask(Pipe):
losses[self.name] += loss
class SimilarityHook(Pipe):
"""
Experimental: A pipeline component to install a hook for supervised
similarity into `Doc` objects. Requires a `Tensorizer` to pre-process
documents. The similarity model can be any object obeying the Thinc `Model`
interface. By default, the model concatenates the elementwise mean and
elementwise max of the two tensors, and compares them using the
Cauchy-like similarity function from Chen (2013):
>>> similarity = 1. / (1. + (W * (vec1-vec2)**2).sum())
Where W is a vector of dimension weights, initialized to 1.
"""
name = 'similarity'
def __init__(self, vocab, model=True, **cfg):
self.vocab = vocab
self.model = model
self.cfg = dict(cfg)
@classmethod
def Model(cls, length):
return Siamese(Pooling(max_pool, mean_pool), CauchySimilarity(length))
def __call__(self, doc):
"""Install similarity hook"""
doc.user_hooks['similarity'] = self.predict
return doc
def pipe(self, docs, **kwargs):
for doc in docs:
yield self(doc)
def predict(self, doc1, doc2):
self.require_model()
return self.model.predict([(doc1, doc2)])
def update(self, doc1_doc2, golds, sgd=None, drop=0.):
self.require_model()
sims, bp_sims = self.model.begin_update(doc1_doc2, drop=drop)
def begin_training(self, _=tuple(), pipeline=None, sgd=None, **kwargs):
"""Allocate model, using width from tensorizer in pipeline.
gold_tuples (iterable): Gold-standard training data.
pipeline (list): The pipeline the model is part of.
"""
if self.model is True:
self.model = self.Model(pipeline[0].model.nO)
link_vectors_to_models(self.vocab)
if sgd is None:
sgd = self.create_optimizer()
return sgd
class TextCategorizer(Pipe):
name = 'textcat'
@ -1299,13 +1014,12 @@ cdef class DependencyParser(Parser):
cdef class EntityRecognizer(Parser):
name = 'ner'
name = "ner"
TransitionSystem = BiluoPushDown
nr_feature = 6
def add_multitask_objective(self, target):
if target == 'cloze':
if target == "cloze":
cloze = ClozeMultitask(self.vocab)
self._multitasks.append(cloze)
else:
@ -1326,8 +1040,8 @@ cdef class EntityRecognizer(Parser):
def labels(self):
# Get the labels from the model by looking at the available moves, e.g.
# B-PERSON, I-PERSON, L-PERSON, U-PERSON
return [move.split('-')[1] for move in self.move_names
if move[0] in ('B', 'I', 'L', 'U')]
return [move.split("-")[1] for move in self.move_names
if move[0] in ("B", "I", "L", "U")]
__all__ = ['Tagger', 'DependencyParser', 'EntityRecognizer', 'Tensorizer', 'TextCategorizer']