|
--- |
|
license: cc-by-sa-4.0 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
- f1 |
|
model-index: |
|
- name: legal-NER |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# legal-NER |
|
|
|
This model is a fine-tuned version of [nlpaueb/legal-bert-base-uncased](https://huggingface.co/nlpaueb/legal-bert-base-uncased) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0068 |
|
- Accuracy: 0.9990 |
|
- Precision: 0.9931 |
|
- Recall: 0.9944 |
|
- F1: 0.9938 |
|
- Classification Report: precision recall f1-score support |
|
|
|
LOC 1.00 1.00 1.00 1837 |
|
MISC 0.98 0.98 0.98 922 |
|
ORG 1.00 0.99 0.99 1341 |
|
PER 1.00 1.00 1.00 1842 |
|
|
|
micro avg 0.99 0.99 0.99 5942 |
|
macro avg 0.99 0.99 0.99 5942 |
|
weighted avg 0.99 0.99 0.99 5942 |
|
|
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Classification Report | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:| |
|
| 0.1501 | 1.0 | 217 | 0.0704 | 0.9810 | 0.8615 | 0.8901 | 0.8756 | precision recall f1-score support |
|
|
|
LOC 0.86 0.95 0.91 1837 |
|
MISC 0.74 0.70 0.72 922 |
|
ORG 0.80 0.82 0.81 1341 |
|
PER 0.97 0.97 0.97 1842 |
|
|
|
micro avg 0.86 0.89 0.88 5942 |
|
macro avg 0.84 0.86 0.85 5942 |
|
weighted avg 0.86 0.89 0.87 5942 |
|
| |
|
| 0.0682 | 2.0 | 434 | 0.0266 | 0.9929 | 0.9513 | 0.9631 | 0.9572 | precision recall f1-score support |
|
|
|
LOC 0.98 0.98 0.98 1837 |
|
MISC 0.88 0.91 0.89 922 |
|
ORG 0.92 0.96 0.94 1341 |
|
PER 0.99 0.97 0.98 1842 |
|
|
|
micro avg 0.95 0.96 0.96 5942 |
|
macro avg 0.94 0.96 0.95 5942 |
|
weighted avg 0.95 0.96 0.96 5942 |
|
| |
|
| 0.0362 | 3.0 | 651 | 0.0137 | 0.9970 | 0.9776 | 0.9850 | 0.9813 | precision recall f1-score support |
|
|
|
LOC 0.98 1.00 0.99 1837 |
|
MISC 0.94 0.95 0.94 922 |
|
ORG 0.98 0.98 0.98 1341 |
|
PER 0.99 1.00 1.00 1842 |
|
|
|
micro avg 0.98 0.99 0.98 5942 |
|
macro avg 0.97 0.98 0.98 5942 |
|
weighted avg 0.98 0.99 0.98 5942 |
|
| |
|
| 0.0209 | 4.0 | 868 | 0.0079 | 0.9986 | 0.9894 | 0.9918 | 0.9906 | precision recall f1-score support |
|
|
|
LOC 0.99 1.00 1.00 1837 |
|
MISC 0.98 0.97 0.97 922 |
|
ORG 0.99 0.99 0.99 1341 |
|
PER 1.00 1.00 1.00 1842 |
|
|
|
micro avg 0.99 0.99 0.99 5942 |
|
macro avg 0.99 0.99 0.99 5942 |
|
weighted avg 0.99 0.99 0.99 5942 |
|
| |
|
| 0.0143 | 5.0 | 1085 | 0.0068 | 0.9990 | 0.9931 | 0.9944 | 0.9938 | precision recall f1-score support |
|
|
|
LOC 1.00 1.00 1.00 1837 |
|
MISC 0.98 0.98 0.98 922 |
|
ORG 1.00 0.99 0.99 1341 |
|
PER 1.00 1.00 1.00 1842 |
|
|
|
micro avg 0.99 0.99 0.99 5942 |
|
macro avg 0.99 0.99 0.99 5942 |
|
weighted avg 0.99 0.99 0.99 5942 |
|
| |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.30.2 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.13.3 |
|
|
