spaCy/website/docs/api/span-predictor.md

title	tag	source	new	teaser	api_base_class	api_string_name	api_trainable
SpanPredictor	class	spacy/pipeline/span_predictor.py	3.4	Pipeline component for resolving tokens into spans	/api/pipe	span_predictor	true

A SpanPredictor component takes in tokens (represented as Spans of length

1. and resolves them into Spans of arbitrary length. The initial use case is as a post-processing step on word-level coreference resolution. The input and output keys used to store Spans are configurable.

Assigned Attributes

Predictions will be saved to Doc.spans as SpanGroups.

Input token spans will be read in using an input prefix, by default "coref_head_clusters", and output spans will be saved using an output prefix (default "coref_clusters") plus a serial number starting from zero. The prefixes are configurable.

Location	Value
Doc.spans[output_prefix + "_" + cluster_number]	One group of predicted spans. SpanGroup

Config and implementation

The default config is defined by the pipeline component factory and describes how the component should be configured. You can override its settings via the config argument on nlp.add_pipe or in your config.cfg for training. See the model architectures documentation for details on the architectures and their arguments and hyperparameters.

Example

from spacy.pipeline.span_predictor import DEFAULT_SPAN_PREDICTOR_MODEL
config={
    "model": DEFAULT_SPAN_PREDICTOR_MODEL,
    "span_cluster_prefix": DEFAULT_CLUSTER_PREFIX,
},
nlp.add_pipe("span_predictor", config=config)

Setting	Description
model	The Model powering the pipeline component. Defaults to SpanPredictor. Model
input_prefix	The prefix to use for input SpanGroups. Defaults to coref_head_clusters. str
output_prefix	The prefix for predicted SpanGroups. Defaults to coref_clusters. str

%%GITHUB_SPACY/spacy/pipeline/span_predictor.py

SpanPredictor.__init__

Example

# Construction via add_pipe with default model
span_predictor = nlp.add_pipe("span_predictor")

# Construction via add_pipe with custom model
config = {"model": {"@architectures": "my_span_predictor.v1"}}
span_predictor = nlp.add_pipe("span_predictor", config=config)

# Construction from class
from spacy.pipeline import SpanPredictor
span_predictor = SpanPredictor(nlp.vocab, model)

Create a new pipeline instance. In your application, you would normally use a shortcut for this and instantiate the component using its string name and nlp.add_pipe.

Name	Description
vocab	The shared vocabulary. Vocab
model	The Model powering the pipeline component. Model
name	String name of the component instance. Used to add entries to the losses during training. str
keyword-only
input_prefix	The prefix to use for input SpanGroups. Defaults to coref_head_clusters. str
output_prefix	The prefix for predicted SpanGroups. Defaults to coref_clusters. str

SpanPredictor.__call__

Apply the pipe to one document. The document is modified in place and returned. This usually happens under the hood when the nlp object is called on a text and all pipeline components are applied to the Doc in order. Both __call__ and pipe delegate to the predict and set_annotations methods.

Example

doc = nlp("This is a sentence.")
span_predictor = nlp.add_pipe("span_predictor")
# This usually happens under the hood
processed = span_predictor(doc)

Name	Description
doc	The document to process. Doc
RETURNS	The processed document. Doc

SpanPredictor.pipe

Apply the pipe to a stream of documents. This usually happens under the hood when the nlp object is called on a text and all pipeline components are applied to the Doc in order. Both __call__ and pipe delegate to the predict and set_annotations methods.

Example

span_predictor = nlp.add_pipe("span_predictor")
for doc in span_predictor.pipe(docs, batch_size=50):
    pass

Name	Description
stream	A stream of documents. Iterable[Doc]
keyword-only
batch_size	The number of documents to buffer. Defaults to 128. int
YIELDS	The processed documents in order. Doc

SpanPredictor.initialize

Initialize the component for training. get_examples should be a function that returns an iterable of Example objects. The data examples are used to initialize the model of the component and can either be the full training data or a representative sample. Initialization includes validating the network, inferring missing shapes and setting up the label scheme based on the data. This method is typically called by Language.initialize.

Example

span_predictor = nlp.add_pipe("span_predictor")
span_predictor.initialize(lambda: [], nlp=nlp)

Name	Description
get_examples	Function that returns gold-standard annotations in the form of Example objects. Callable], Iterable[Example
keyword-only
nlp	The current nlp object. Defaults to None. Optional[Language]

SpanPredictor.predict

Apply the component's model to a batch of Doc objects, without modifying them. Predictions are returned as a list of MentionClusters, one for each input Doc. A MentionClusters instance is just a list of lists of pairs of ints, where each item corresponds to an input SpanGroup, and the ints correspond to token indices.

Example

span_predictor = nlp.add_pipe("span_predictor")
spans = span_predictor.predict([doc1, doc2])

Name	Description
docs	The documents to predict. Iterable[Doc]
RETURNS	The predicted spans for the Docs. List[MentionClusters]

SpanPredictor.set_annotations

Modify a batch of documents, saving predictions using the output prefix in Doc.spans.

Example

span_predictor = nlp.add_pipe("span_predictor")
spans = span_predictor.predict([doc1, doc2])
span_predictor.set_annotations([doc1, doc2], spans)

Name	Description
docs	The documents to modify. Iterable[Doc]
spans	The predicted spans for the docs. List[MentionClusters]

SpanPredictor.update

Learn from a batch of Example objects. Delegates to predict.

Example

span_predictor = nlp.add_pipe("span_predictor")
optimizer = nlp.initialize()
losses = span_predictor.update(examples, sgd=optimizer)

Name	Description
examples	A batch of Example objects to learn from. Iterable[Example]
keyword-only
drop	The dropout rate. float
sgd	An optimizer. Will be created via create_optimizer if not set. Optional[Optimizer]
losses	Optional record of the loss during training. Updated using the component name as the key. Optional[Dict[str, float]]
RETURNS	The updated losses dictionary. Dict[str, float]

SpanPredictor.create_optimizer

Create an optimizer for the pipeline component.

Example

span_predictor = nlp.add_pipe("span_predictor")
optimizer = span_predictor.create_optimizer()

Name	Description
RETURNS	The optimizer. Optimizer

SpanPredictor.use_params

Modify the pipe's model, to use the given parameter values. At the end of the context, the original parameters are restored.

Example

span_predictor = nlp.add_pipe("span_predictor")
with span_predictor.use_params(optimizer.averages):
    span_predictor.to_disk("/best_model")

Name	Description
params	The parameter values to use in the model. dict

SpanPredictor.to_disk

Serialize the pipe to disk.

Example

span_predictor = nlp.add_pipe("span_predictor")
span_predictor.to_disk("/path/to/span_predictor")

Name	Description
path	A path to a directory, which will be created if it doesn't exist. Paths may be either strings or Path-like objects. Union[str, Path]
keyword-only
exclude	String names of serialization fields to exclude. Iterable[str]

SpanPredictor.from_disk

Load the pipe from disk. Modifies the object in place and returns it.

Example

span_predictor = nlp.add_pipe("span_predictor")
span_predictor.from_disk("/path/to/span_predictor")

Name	Description
path	A path to a directory. Paths may be either strings or Path-like objects. Union[str, Path]
keyword-only
exclude	String names of serialization fields to exclude. Iterable[str]
RETURNS	The modified SpanPredictor object. SpanPredictor

SpanPredictor.to_bytes

Example

span_predictor = nlp.add_pipe("span_predictor")
span_predictor_bytes = span_predictor.to_bytes()

Serialize the pipe to a bytestring.

Name	Description
keyword-only
exclude	String names of serialization fields to exclude. Iterable[str]
RETURNS	The serialized form of the SpanPredictor object. bytes

SpanPredictor.from_bytes

Load the pipe from a bytestring. Modifies the object in place and returns it.

Example

span_predictor_bytes = span_predictor.to_bytes()
span_predictor = nlp.add_pipe("span_predictor")
span_predictor.from_bytes(span_predictor_bytes)

Name	Description
bytes_data	The data to load from. bytes
keyword-only
exclude	String names of serialization fields to exclude. Iterable[str]
RETURNS	The SpanPredictor object. SpanPredictor

Serialization fields

During serialization, spaCy will export several data fields used to restore different aspects of the object. If needed, you can exclude them from serialization by passing in the string names via the exclude argument.

Example

data = span_predictor.to_disk("/path", exclude=["vocab"])

Name	Description
vocab	The shared Vocab.
cfg	The config file. You usually don't want to exclude this.
model	The binary model data. You usually don't want to exclude this.