训练spaCy SpanCat模型遇E143标签未初始化错误，求解决方案

问题详情 #

训练spaCy SpanCat模型时触发错误：

ValueError: [E143] Labels for component 'spancat' not initialized. This can be fixed by calling add_label, or by providing a representative batch of examples to the component's 'initialize' method.

已将NER ents转换为spans，转换代码：

def main(loc: Path, lang: str, span_key: str):
    """
    Set the NER data into the doc.spans, under a given key.
    The SpanCategorizer component uses the doc.spans, so that it can work with
    overlapping or nested annotations, which can't be represented on the
    per-token level.
    """
    nlp = spacy.blank(lang)
    docbin = DocBin().from_disk(loc)
    docs = list(docbin.get_docs(nlp.vocab))
    for doc in docs:
        doc.spans[span_key] = list(doc.ents)
    DocBin(docs=docs).to_disk(loc)

使用的配置文件：

[paths]
train = null
dev = null
vectors = null
init_tok2vec = null

[system]
gpu_allocator = null
seed = 444

[nlp]
lang = "en"
pipeline = ["tok2vec","spancat"]
batch_size = 1000
disabled = []
before_creation = null
after_creation = null
after_pipeline_creation = null
tokenizer = {"@tokenizers":"spacy.Tokenizer.v1"}

[components]

[components.spancat]
factory = "spancat"
max_positive = null
scorer = {"@scorers":"spacy.spancat_scorer.v1"}
spans_key = "sc"
threshold = 0.5

[components.spancat.model]
@architectures = "spacy.SpanCategorizer.v1"

[components.spancat.model.reducer]
@layers = "spacy.mean_max_reducer.v1"
hidden_size = 128

[components.spancat.model.scorer]
@layers = "spacy.LinearLogistic.v1"
nO = null
nI = null

[components.spancat.model.tok2vec]
@architectures = "spacy.Tok2VecListener.v1"
width = ${components.tok2vec.model.encode.width}
upstream = "*"

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

[components.tok2vec]
factory = "tok2vec"

[components.tok2vec.model]
@architectures = "spacy.Tok2Vec.v2"

[components.tok2vec.model.embed]
@architectures = "spacy.MultiHashEmbed.v2"
width = ${components.tok2vec.model.encode.width}
attrs = ["NORM","PREFIX","SUFFIX","SHAPE"]
rows = [5000,1000,2500,2500]
include_static_vectors = true

[components.tok2vec.model.encode]
@architectures = "spacy.MaxoutWindowEncoder.v2"
width = 256
depth = 8
window_size = 1
maxout_pieces = 3

[corpora]

[corpora.dev]
@readers = "spacy.Corpus.v1"
path = ${paths.dev}
max_length = 0
gold_preproc = false
limit = 0
augmenter = null

[corpora.train]
@readers = "spacy.Corpus.v1"
path = ${paths.train}
max_length = 0
gold_preproc = false
limit = 0
augmenter = null

[training]
dev_corpus = "corpora.dev"
train_corpus = "corpora.train"
max_epochs = 70
seed = ${system.seed}
gpu_allocator = ${system.gpu_allocator}
dropout = 0.1
accumulate_gradient = 1
patience = 1600
max_steps = 20000
eval_frequency = 200
frozen_components = []
annotating_components = []
before_to_disk = null

[training.batcher]
@batchers = "spacy.batch_by_words.v1"
discard_oversize = false
tolerance = 0.2
get_length = null

[training.batcher.size]
@schedules = "compounding.v1"
start = 100
stop = 1000
compound = 1.001
t = 0.0

[training.logger]
@loggers = "spacy.ConsoleLogger.v1"
progress_bar = false

[training.optimizer]
@optimizers = "Adam.v1"
beta1 = 0.9
beta2 = 0.999
L2_is_weight_decay = true
L2 = 0.01
grad_clip = 1.0
use_averages = false
eps = 0.00000001
learn_rate = 0.001

[training.score_weights]
spans_sc_f = 1.0
spans_sc_p = 0.0
spans_sc_r = 0.0

[pretraining]

[initialize]
vectors = ${paths.vectors}
init_tok2vec = ${paths.init_tok2vec}
vocab_data = null
lookups = null
before_init = null
after_init = null

[initialize.components]

[initialize.tokenizer]

使用的span键为"sc"。

解决方案 #

方法一：配置自动从训练数据初始化标签（推荐） #

修改配置文件的[initialize.components]部分，添加spancat的初始化设置，让spaCy自动从训练数据的doc.spans["sc"]中提取标签：

[initialize.components]

[initialize.components.spancat]
@initialize = "spacy.initialize_spancat.v1"
spans_key = "sc"

添加后，spaCy在初始化阶段会自动扫描训练数据，获取所有span标签并完成组件初始化。

方法二：手动添加标签（适用于已知所有标签的场景） #

如果已经明确所有需要识别的span标签，可以在代码中手动添加：

import spacy
from spacy.tokens import DocBin

# 加载配置并禁用spancat组件先
nlp = spacy.load("your_config.cfg", disable=["spancat"])
# 添加spancat组件，指定span_key
spancat = nlp.add_pipe("spancat", config={"spans_key": "sc"})

# 替换为你的实际标签列表，比如从原NER数据中获取的标签
labels = ["PERSON", "ORG", "LOC"]
for label in labels:
    spancat.add_label(label)

# 读取训练数据用于初始化
docbin = DocBin().from_disk("your_train_data_path")
train_docs = list(docbin.get_docs(nlp.vocab))

# 初始化组件
spacy.util.initialize_bigram_suggester(nlp.vocab, sizes=[1,2,3])
spancat.initialize(lambda: nlp.pipe(train_docs), nlp=nlp)

# 保存初始化后的模型或直接开始训练
nlp.to_disk("initialized_model")

额外验证步骤 #

确保转换后的数据集确实包含正确的span标签：

import spacy
from spacy.tokens import DocBin

nlp = spacy.blank("en")
docbin = DocBin().from_disk("your_train_data_path")
docs = list(docbin.get_docs(nlp.vocab))

# 打印前5个文档的span信息，确认标签存在
for doc in docs[:5]:
    spans = doc.spans.get("sc", [])
    print(f"文档内容: {doc.text}")
    print(f"Span标签: {[(span.text, span.label_) for span in spans]}")

如果输出中没有标签或spans为空，说明数据转换步骤存在问题，需要检查原NER数据是否正确。

内容的提问来源于stack exchange，提问作者SteveS