metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- nubes
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: bsc-finetuned-ner
results:
- task:
name: Token Classification
type: token-classification
dataset:
name: nubes
type: nubes
config: Nubes
split: train
args: Nubes
metrics:
- name: Precision
type: precision
value: 0.8846931894807822
- name: Recall
type: recall
value: 0.9190893169877408
- name: F1
type: f1
value: 0.9015633052740079
- name: Accuracy
type: accuracy
value: 0.9765241847383958
bsc-finetuned-ner
This model is a fine-tuned version of PlanTL-GOB-ES/bsc-bio-ehr-es on the nubes dataset. It achieves the following results on the evaluation set:
- Loss: 0.1548
- Precision: 0.8847
- Recall: 0.9191
- F1: 0.9016
- Accuracy: 0.9765
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 7
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1179 | 1.0 | 1726 | 0.1078 | 0.8223 | 0.8673 | 0.8442 | 0.9682 |
| 0.0775 | 2.0 | 3452 | 0.1152 | 0.8474 | 0.8872 | 0.8669 | 0.9722 |
| 0.0424 | 3.0 | 5178 | 0.1096 | 0.8677 | 0.9054 | 0.8862 | 0.9753 |
| 0.023 | 4.0 | 6904 | 0.1301 | 0.8740 | 0.9040 | 0.8888 | 0.9753 |
| 0.0098 | 5.0 | 8630 | 0.1352 | 0.8829 | 0.9194 | 0.9008 | 0.9778 |
| 0.0088 | 6.0 | 10356 | 0.1483 | 0.8903 | 0.9121 | 0.9010 | 0.9756 |
| 0.0045 | 7.0 | 12082 | 0.1548 | 0.8847 | 0.9191 | 0.9016 | 0.9765 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2