metadata
license: mit
base_model: PORTULAN/albertina-100m-portuguese-ptpt-encoder
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: LVI_albertina-100m-portuguese-ptpt-encoder
results: []
LVI_albertina-100m-portuguese-ptpt-encoder
This model is a fine-tuned version of PORTULAN/albertina-100m-portuguese-ptpt-encoder on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1867
- Accuracy: 0.9802
- F1: 0.9800
- Precision: 0.9905
- Recall: 0.9696
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-06
- 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.1284 | 1.0 | 3217 | 0.1454 | 0.9581 | 0.9567 | 0.9882 | 0.9272 |
| 0.0946 | 2.0 | 6434 | 0.1211 | 0.9737 | 0.9734 | 0.9864 | 0.9607 |
| 0.0575 | 3.0 | 9651 | 0.1087 | 0.9776 | 0.9774 | 0.9892 | 0.9659 |
| 0.0374 | 4.0 | 12868 | 0.1033 | 0.981 | 0.9809 | 0.9854 | 0.9765 |
| 0.0311 | 5.0 | 16085 | 0.1154 | 0.981 | 0.9808 | 0.9896 | 0.9722 |
| 0.0125 | 6.0 | 19302 | 0.1143 | 0.9830 | 0.9830 | 0.9833 | 0.9826 |
| 0.0107 | 7.0 | 22519 | 0.1562 | 0.9807 | 0.9805 | 0.9910 | 0.9702 |
| 0.0032 | 8.0 | 25736 | 0.1711 | 0.9808 | 0.9806 | 0.9892 | 0.9721 |
| 0.0036 | 9.0 | 28953 | 0.1867 | 0.9802 | 0.9800 | 0.9905 | 0.9696 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2