pipeline_tag: 标记分类 tags:

• 命名实体识别
• 序列标注模型 widget:
• text: 我的名字是Amadeus Wolfgang，居住在柏林 inference: parameters: aggregation_strategy: simple grouped_entities: true language:
• ro

基于xlm-roberta模型在ronec数据集上训练，测试集f1-Macro达到95分。

from transformers import AutoTokenizer, AutoModelForTokenClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("EvanD/xlm-roberta-base-romanian-ner-ronec")
ner_model = AutoModelForTokenClassification.from_pretrained("EvanD/xlm-roberta-base-romanian-ner-ronec")

nlp = pipeline("ner", model=ner_model, tokenizer=tokenizer, aggregation_strategy="simple")
example = "我的名字是Amadeus Wolfgang，居住在柏林"

ner_results = nlp(example)
print(ner_results)

# [
#     {
#         'entity_group': 'PER',
#         'score': 0.9966806,
#         'word': 'Amadeus Wolfgang',
#         'start': 6,
#         'end': 22
#     },
#     {'entity_group': 'GPE',
#      'score': 0.99694663,
#      'word': '柏林',
#      'start': 27,
#      'end': 29
#      }
# ]