metadata
base_model: dccuchile/bert-base-spanish-wwm-cased
tags:
- generated_from_trainer
metrics:
- f1
- recall
model-index:
- name: bert-base-spanish-wwm-cased_K5
results: []
bert-base-spanish-wwm-cased_K5
This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0182
- F1 Macro: 0.9973
- F1: 0.9980
- F1 Neg: 0.9966
- Acc: 0.9975
- Prec: 0.9980
- Recall: 0.9980
- Mcc: 0.9946
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: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 | F1 Neg | Acc | Prec | Recall | Mcc |
|---|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 400 | 0.0263 | 0.9960 | 0.9970 | 0.9949 | 0.9962 | 0.9980 | 0.9961 | 0.9919 |
| 0.0591 | 2.0 | 800 | 0.0182 | 0.9973 | 0.9980 | 0.9966 | 0.9975 | 0.9980 | 0.9980 | 0.9946 |
| 0.0123 | 3.0 | 1200 | 0.0225 | 0.9973 | 0.9980 | 0.9966 | 0.9975 | 0.9980 | 0.9980 | 0.9946 |
| 0.0078 | 4.0 | 1600 | 0.0227 | 0.9973 | 0.9980 | 0.9966 | 0.9975 | 0.9980 | 0.9980 | 0.9946 |
| 0.002 | 5.0 | 2000 | 0.0229 | 0.9973 | 0.9980 | 0.9966 | 0.9975 | 0.9980 | 0.9980 | 0.9946 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2