metadata

license: cc-by-sa-4.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: legal-bert-small-NER
    results: []

legal-bert-small-NER

This model is a fine-tuned version of nlpaueb/legal-bert-small-uncased on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.2334
Accuracy: 0.9558
Precision: 0.7587
Recall: 0.7950
F1: 0.7764

Classification Report: precision recall f1-score support

   LOC       0.85      0.86      0.86      1668
  MISC       0.56      0.67      0.61       702
   ORG       0.68      0.67      0.68      1661
   PER       0.83      0.91      0.87      1617

micro avg 0.76 0.79 0.78 5648 macro avg 0.73 0.78 0.75 5648

weighted avg 0.76 0.79 0.78 5648

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 16
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Classification Report
0.0289	1.0	434	0.2151	0.9555	0.7592	0.7890	0.7738	precision recall f1-score support

     LOC       0.86      0.85      0.86      1668
    MISC       0.57      0.67      0.62       702
     ORG       0.69      0.64      0.67      1661
     PER       0.81      0.92      0.86      1617

micro avg 0.76 0.79 0.77 5648 macro avg 0.73 0.77 0.75 5648 weighted avg 0.76 0.79 0.77 5648 | | 0.0193 | 2.0 | 868 | 0.2334 | 0.9558 | 0.7587 | 0.7950 | 0.7764 | precision recall f1-score support

     LOC       0.85      0.86      0.86      1668
    MISC       0.56      0.67      0.61       702
     ORG       0.68      0.67      0.68      1661
     PER       0.83      0.91      0.87      1617

micro avg 0.76 0.79 0.78 5648 macro avg 0.73 0.78 0.75 5648 weighted avg 0.76 0.79 0.78 5648 |

Framework versions

Transformers 4.30.2
Pytorch 2.0.0
Datasets 2.1.0
Tokenizers 0.13.3