metadata

datasets:
  - wnut2017
metrics:
  - f1
  - precision
  - recall
model-index:
  - name: tner/bertweet-large-wnut2017
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: wnut2017
          type: wnut2017
          args: wnut2017
        metrics:
          - name: F1
            type: f1
            value: 0.5302273987798114
          - name: Precision
            type: precision
            value: 0.6602209944751382
          - name: Recall
            type: recall
            value: 0.44300278035217794
          - name: F1 (macro)
            type: f1_macro
            value: 0.4643459997680019
          - name: Precision (macro)
            type: precision_macro
            value: 0.5792841925426832
          - name: Recall (macro)
            type: recall_macro
            value: 0.3973128655628379
          - name: F1 (entity span)
            type: f1_entity_span
            value: 0.6142697881828317
          - name: Precision (entity span)
            type: precision_entity_span
            value: 0.7706293706293706
          - name: Recall (entity span)
            type: recall_entity_span
            value: 0.5106580166821131
pipeline_tag: token-classification
widget:
  - text: Jacob Collier is a Grammy awarded artist from England.
    example_title: NER Example 1

tner/bertweet-large-wnut2017

This model is a fine-tuned version of vinai/bertweet-large on the tner/wnut2017 dataset. Model fine-tuning is done via T-NER's hyper-parameter search (see the repository for more detail). It achieves the following results on the test set:

F1 (micro): 0.5302273987798114
Precision (micro): 0.6602209944751382
Recall (micro): 0.44300278035217794
F1 (macro): 0.4643459997680019
Precision (macro): 0.5792841925426832
Recall (macro): 0.3973128655628379

The per-entity breakdown of the F1 score on the test set are below:

corporation: 0.3902439024390244
group: 0.37130801687763715
location: 0.6595744680851063
person: 0.65474552957359
product: 0.2857142857142857
work_of_art: 0.4244897959183674

For F1 scores, the confidence interval is obtained by bootstrap as below:

F1 (micro):
- 90%: [0.5002577319587629, 0.5587481638299118]
- 95%: [0.4947163587619384, 0.5629013150503995]
F1 (macro):
- 90%: [0.5002577319587629, 0.5587481638299118]
- 95%: [0.4947163587619384, 0.5629013150503995]

Full evaluation can be found at metric file of NER and metric file of entity span.

Training hyperparameters

The following hyperparameters were used during training:

dataset: ['tner/wnut2017']
dataset_split: train
dataset_name: None
local_dataset: None
model: vinai/bertweet-large
crf: False
max_length: 128
epoch: 15
batch_size: 16
lr: 1e-05
random_seed: 42
gradient_accumulation_steps: 4
weight_decay: 1e-07
lr_warmup_step_ratio: 0.1
max_grad_norm: 10.0

The full configuration can be found at fine-tuning parameter file.

Reference

If you use any resource from T-NER, please consider to cite our 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.",
}