metadata
language:
- es
tags:
- generated_from_trainer
datasets:
- jorgeortizfuentes/spanish_nominal_groups_conll2003
model-index:
- name: nominal-groups-recognition-bert-base-spanish-wwm-cased
results: []
nominal-groups-recognition-bert-base-spanish-wwm-cased
This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on the jorgeortizfuentes/spanish_nominal_groups_conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2988
- Ng Precision: 0.7309
- Ng Recall: 0.7780
- Ng F1: 0.7537
- Ng Number: 3198
- Overall Precision: 0.7309
- Overall Recall: 0.7780
- Overall F1: 0.7537
- Overall Accuracy: 0.9019
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: 16
- seed: 13
- 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 | Ng Precision | Ng Recall | Ng F1 | Ng Number | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.451 | 1.0 | 114 | 0.2882 | 0.6985 | 0.7483 | 0.7225 | 3198 | 0.6985 | 0.7483 | 0.7225 | 0.8899 |
| 0.2429 | 2.0 | 228 | 0.2917 | 0.7294 | 0.7483 | 0.7387 | 3198 | 0.7294 | 0.7483 | 0.7387 | 0.8932 |
| 0.193 | 3.0 | 342 | 0.2864 | 0.7306 | 0.7727 | 0.7511 | 3198 | 0.7306 | 0.7727 | 0.7511 | 0.9000 |
| 0.1586 | 4.0 | 456 | 0.2988 | 0.7309 | 0.7780 | 0.7537 | 3198 | 0.7309 | 0.7780 | 0.7537 | 0.9019 |
| 0.1386 | 5.0 | 570 | 0.3116 | 0.7275 | 0.7770 | 0.7514 | 3198 | 0.7275 | 0.7770 | 0.7514 | 0.9003 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3