tner/roberta-large-ontonotes5

这个模型是在 tner/ontonotes5 数据集上通过 T-NER 的超参数搜索进行的fine-tuned版本。该模型在测试集上达到以下结果：

• F1 (micro): 0.908632361399938
• Precision (micro): 0.905148095909732
• Recall (micro): 0.9121435551212579
• F1 (macro): 0.8265477704565624
• Precision (macro): 0.8170668848546687
• Recall (macro): 0.8387672780349001

对于测试集上的F1分数的每个实体的详细情况如下：

• cardinal_number: 0.8605277329025309
• date: 0.872996300863132
• event: 0.7424242424242424
• facility: 0.7732342007434945
• geopolitical_area: 0.9687148323205043
• group: 0.9470588235294117
• language: 0.7499999999999999
• law: 0.6666666666666666
• location: 0.7593582887700535
• money: 0.901098901098901
• ordinal_number: 0.85785536159601
• organization: 0.9227360841872057
• percent: 0.9171428571428571
• person: 0.9556004036326943
• product: 0.7857142857142858
• quantity: 0.7945205479452055
• time: 0.6870588235294116
• work_of_art: 0.7151515151515151

对于F1分数，置信区间通过自助法获取如下：

• F1 (micro):
  • 90%: [0.9039454247544766, 0.9128956119702822]
  • 95%: [0.9030263216115454, 0.9138350859566045]
• F1 (macro):
  • 90%: [0.9039454247544766, 0.9128956119702822]
  • 95%: [0.9030263216115454, 0.9138350859566045]

完整的评估可以在 metric file of NER 和 metric file of entity span 找到。

使用方法

可以通过 tner library 使用该模型。使用pip安装该库

pip install tner

并按照以下方式激活模型。

from tner import TransformersNER
model = TransformersNER("tner/roberta-large-ontonotes5")
model.predict(["Jacob Collier is a Grammy awarded English artist from London"])

可以通过transformers库使用它，但目前不推荐，因为不支持CRF层。

训练超参数

训练过程中使用了以下超参数：

• dataset: ['tner/ontonotes5']
• dataset_split: train
• dataset_name: None
• local_dataset: None
• model: roberta-large
• crf: True
• max_length: 128
• epoch: 15
• batch_size: 64
• lr: 1e-05
• random_seed: 42
• gradient_accumulation_steps: 1
• weight_decay: None
• lr_warmup_step_ratio: 0.1
• max_grad_norm: 10.0

完整的配置可以在 fine-tuning parameter file 找到。

参考

如果您使用了T-NER的任何资源，请考虑引用我们的 paper 。

@inproceedings{ushio-camacho-collados-2021-ner,
    title = "{T}-{NER}: An All-Round Python Library for Transformer-based Named Entity Recognition",
    author = "Ushio, Asahi  and
      Camacho-Collados, Jose",
    booktitle = "Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations",
    month = apr,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.eacl-demos.7",
    doi = "10.18653/v1/2021.eacl-demos.7",
    pages = "53--62",
    abstract = "Language model (LM) pretraining has led to consistent improvements in many NLP downstream tasks, including named entity recognition (NER). In this paper, we present T-NER (Transformer-based Named Entity Recognition), a Python library for NER LM finetuning. In addition to its practical utility, T-NER facilitates the study and investigation of the cross-domain and cross-lingual generalization ability of LMs finetuned on NER. Our library also provides a web app where users can get model predictions interactively for arbitrary text, which facilitates qualitative model evaluation for non-expert programmers. We show the potential of the library by compiling nine public NER datasets into a unified format and evaluating the cross-domain and cross- lingual performance across the datasets. The results from our initial experiments show that in-domain performance is generally competitive across datasets. However, cross-domain generalization is challenging even with a large pretrained LM, which has nevertheless capacity to learn domain-specific features if fine- tuned on a combined dataset. To facilitate future research, we also release all our LM checkpoints via the Hugging Face model hub.",
}