SpanCategorizer

A span categorizer consists of two parts: a suggester function that proposes candidate spans, which may or may not overlap, and a labeler model that predicts zero or more labels for each candidate.

This component comes in two forms: spancat and spancat_singlelabel (added in spaCy v3.5.1). When you need to perform multi-label classification on your spans, use spancat. The spancat component uses a Logistic layer where the output class probabilities are independent for each class. However, if you need to predict at most one true class for a span, then use spancat_singlelabel. It uses a Softmax layer and treats the task as a multi-class problem.

Predicted spans will be saved in a SpanGroup on the doc under doc.spans[spans_key], where spans_key is a component config setting. Individual span scores are stored in doc.spans[spans_key].attrs["scores"].

Assigned Attributes {id="assigned-attributes"}

Predictions will be saved to Doc.spans[spans_key] as a SpanGroup. The scores for the spans in the SpanGroup will be saved in SpanGroup.attrs["scores"].

spans_key defaults to "sc", but can be passed as a parameter. The spancat component will overwrite any existing spans under the spans key doc.spans[spans_key].

Config and implementation {id="config"}

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 (spancat)

pythonCopy

from spacy.pipeline.spancat import DEFAULT_SPANCAT_MODEL
config = {
    "threshold": 0.5,
    "spans_key": "labeled_spans",
    "max_positive": None,
    "model": DEFAULT_SPANCAT_MODEL,
    "suggester": {"@misc": "spacy.ngram_suggester.v1", "sizes": [1, 2, 3]},
}
nlp.add_pipe("spancat", config=config)

Example (spancat_singlelabel)

pythonCopy

from spacy.pipeline.spancat import DEFAULT_SPANCAT_SINGLELABEL_MODEL
config = {
    "spans_key": "labeled_spans",
    "model": DEFAULT_SPANCAT_SINGLELABEL_MODEL,
    "suggester": {"@misc": "spacy.ngram_suggester.v1", "sizes": [1, 2, 3]},
    # Additional spancat_singlelabel parameters
    "negative_weight": 0.8,
    "allow_overlap": True,
}
nlp.add_pipe("spancat_singlelabel", config=config)

If you set a non-default value for spans_key, you'll have to update [training.score_weights] as well so that weights are computed properly. E. g. for spans_key == "myspankey", include this in your config:

[training.score_weights]
spans_myspankey_f = 1.0
spans_myspankey_p = 0.0
spans_myspankey_r = 0.0

%%GITHUB_SPACY/spacy/pipeline/spancat.py

SpanCategorizer.__init__ {id="init",tag="method"}

Example

pythonCopy

# Construction via add_pipe with default model
# Replace 'spancat' with 'spancat_singlelabel' for exclusive classes
spancat = nlp.add_pipe("spancat")

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

# Construction from class
from spacy.pipeline import SpanCategorizer
spancat = SpanCategorizer(nlp.vocab, model, suggester)

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.

SpanCategorizer.__call__ {id="call",tag="method"}

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

pythonCopy

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

SpanCategorizer.pipe {id="pipe",tag="method"}

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

pythonCopy

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

SpanCategorizer.initialize {id="initialize",tag="method"}

Initialize the component for training. get_examples should be a function that returns an iterable of Example objects. At least one example should be supplied. 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 and lets you customize arguments it receives via the [initialize.components] block in the config.

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
spancat.initialize(lambda: examples, nlp=nlp)

iniCopy

### config.cfg
[initialize.components.spancat]

[initialize.components.spancat.labels]
@readers = "spacy.read_labels.v1"
path = "corpus/labels/spancat.json

SpanCategorizer.predict {id="predict",tag="method"}

Apply the component's model to a batch of Doc objects without modifying them.

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
scores = spancat.predict([doc1, doc2])

SpanCategorizer.set_annotations {id="set_annotations",tag="method"}

Modify a batch of Doc objects using pre-computed scores.

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
scores = spancat.predict(docs)
spancat.set_annotations(docs, scores)

SpanCategorizer.update {id="update",tag="method"}

Learn from a batch of Example objects containing the predictions and gold-standard annotations, and update the component's model. Delegates to predict and get_loss.

Example

pythonCopy

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

SpanCategorizer.set_candidates {id="set_candidates",tag="method", version="3.3"}

Use the suggester to add a list of Span candidates to a list of Doc objects. This method is intended to be used for debugging purposes.

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
spancat.set_candidates(docs, "candidates")

SpanCategorizer.get_loss {id="get_loss",tag="method"}

Find the loss and gradient of loss for the batch of documents and their predicted scores.

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
scores = spancat.predict([eg.predicted for eg in examples])
loss, d_loss = spancat.get_loss(examples, scores)

SpanCategorizer.create_optimizer {id="create_optimizer",tag="method"}

Create an optimizer for the pipeline component.

Example

pythonCopy

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

SpanCategorizer.use_params {id="use_params",tag="method, contextmanager"}

Modify the pipe's model to use the given parameter values.

Example

pythonCopy

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

SpanCategorizer.add_label {id="add_label",tag="method"}

Add a new label to the pipe. Raises an error if the output dimension is already set, or if the model has already been fully initialized. Note that you don't have to call this method if you provide a representative data sample to the initialize method. In this case, all labels found in the sample will be automatically added to the model, and the output dimension will be inferred automatically.

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
spancat.add_label("MY_LABEL")

SpanCategorizer.to_disk {id="to_disk",tag="method"}

Serialize the pipe to disk.

Example

pythonCopy

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

SpanCategorizer.from_disk {id="from_disk",tag="method"}

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

Example

pythonCopy

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

SpanCategorizer.to_bytes {id="to_bytes",tag="method"}

Example

pythonCopy

spancat = nlp.add_pipe("spancat")
spancat_bytes = spancat.to_bytes()

Serialize the pipe to a bytestring.

SpanCategorizer.from_bytes {id="from_bytes",tag="method"}

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

Example

pythonCopy

spancat_bytes = spancat.to_bytes()
spancat = nlp.add_pipe("spancat")
spancat.from_bytes(spancat_bytes)

SpanCategorizer.labels {id="labels",tag="property"}

The labels currently added to the component.

Example

pythonCopy

spancat.add_label("MY_LABEL")
assert "MY_LABEL" in spancat.labels

SpanCategorizer.label_data {id="label_data",tag="property"}

The labels currently added to the component and their internal meta information. This is the data generated by init labels and used by SpanCategorizer.initialize to initialize the model with a pre-defined label set.

Example

pythonCopy

labels = spancat.label_data
spancat.initialize(lambda: [], nlp=nlp, labels=labels)

Serialization fields {id="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

pythonCopy

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

Suggesters {id="suggesters",tag="registered functions",source="spacy/pipeline/spancat.py"}

spacy.ngram_suggester.v1 {id="ngram_suggester"}

Example Config

iniCopy

[components.spancat.suggester]
@misc = "spacy.ngram_suggester.v1"
sizes = [1, 2, 3]

Suggest all spans of the given lengths. Spans are returned as a ragged array of integers. The array has two columns, indicating the start and end position.

spacy.ngram_range_suggester.v1 {id="ngram_range_suggester"}

Example Config

iniCopy

[components.spancat.suggester]
@misc = "spacy.ngram_range_suggester.v1"
min_size = 2
max_size = 4

Suggest all spans of at least length min_size and at most length max_size (both inclusive). Spans are returned as a ragged array of integers. The array has two columns, indicating the start and end position.

spacy.preset_spans_suggester.v1 {id="preset_spans_suggester"}

Example Config

iniCopy

[components.spancat.suggester]
@misc = "spacy.preset_spans_suggester.v1"
spans_key = "my_spans"

Suggest all spans that are already stored in doc.spans[spans_key]. This is useful when an upstream component is used to set the spans on the Doc such as a SpanRuler or SpanFinder.