metadata

datasets:
  - tner/fin
metrics:
  - f1
  - precision
  - recall
pipeline_tag: token-classification
widget:
  - text: Jacob Collier is a Grammy awarded artist from England.
    example_title: NER Example 1
base_model: microsoft/deberta-v3-large
model-index:
  - name: tner/deberta-v3-large-fin
    results:
      - task:
          type: token-classification
          name: Token Classification
        dataset:
          name: tner/fin
          type: tner/fin
          args: tner/fin
        metrics:
          - type: f1
            value: 0.7060755336617406
            name: F1
          - type: precision
            value: 0.738831615120275
            name: Precision
          - type: recall
            value: 0.6761006289308176
            name: Recall
          - type: f1_macro
            value: 0.45092058848834204
            name: F1 (macro)
          - type: precision_macro
            value: 0.45426465258085835
            name: Precision (macro)
          - type: recall_macro
            value: 0.45582773707773705
            name: Recall (macro)
          - type: f1_entity_span
            value: 0.7293729372937293
            name: F1 (entity span)
          - type: precision_entity_span
            value: 0.7594501718213058
            name: Precision (entity span)
          - type: recall_entity_span
            value: 0.7015873015873015
            name: Recall (entity span)

tner/deberta-v3-large-fin

This model is a fine-tuned version of microsoft/deberta-v3-large on the tner/fin 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): 0.7060755336617406
Precision (micro): 0.738831615120275
Recall (micro): 0.6761006289308176
F1 (macro): 0.45092058848834204
Precision (macro): 0.45426465258085835
Recall (macro): 0.45582773707773705

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

location: 0.4000000000000001
organization: 0.5762711864406779
other: 0.0
person: 0.8274111675126904

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

F1 (micro):
- 90%: [0.6370316240330781, 0.7718233002182738]
- 95%: [0.6236274300363168, 0.7857205513784461]
F1 (macro):
- 90%: [0.6370316240330781, 0.7718233002182738]
- 95%: [0.6236274300363168, 0.7857205513784461]

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

Usage

This model can be used through the tner library. Install the library via pip

pip install tner

and activate model as below.

from tner import TransformersNER
model = TransformersNER("tner/deberta-v3-large-fin")
model.predict(["Jacob Collier is a Grammy awarded English artist from London"])

It can be used via transformers library but it is not recommended as CRF layer is not supported at the moment.

Training hyperparameters

The following hyperparameters were used during training:

dataset: ['tner/fin']
dataset_split: train
dataset_name: None
local_dataset: None
model: microsoft/deberta-v3-large
crf: True
max_length: 128
epoch: 15
batch_size: 16
lr: 1e-05
random_seed: 42
gradient_accumulation_steps: 4
weight_decay: None
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.",
}