|
--- |
|
license: mit |
|
base_model: neuralmind/bert-base-portuguese-cased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: finetuned-bert-categories-estimation |
|
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. --> |
|
|
|
# finetuned-bert-categories-estimation |
|
|
|
This model is a fine-tuned version of [neuralmind/bert-base-portuguese-cased](https://huggingface.co/neuralmind/bert-base-portuguese-cased) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3765 |
|
- F1: 0.8829 |
|
- Accuracy: 0.9185 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:| |
|
| 4.1819 | 0.13 | 100 | 3.4072 | 0.1568 | 0.4170 | |
|
| 3.0332 | 0.27 | 200 | 2.4751 | 0.3097 | 0.5559 | |
|
| 2.3356 | 0.4 | 300 | 1.9235 | 0.4379 | 0.6605 | |
|
| 1.8533 | 0.53 | 400 | 1.5348 | 0.5528 | 0.7300 | |
|
| 1.5404 | 0.66 | 500 | 1.2910 | 0.6279 | 0.7834 | |
|
| 1.3375 | 0.8 | 600 | 1.0888 | 0.6428 | 0.7987 | |
|
| 1.09 | 0.93 | 700 | 0.9613 | 0.6828 | 0.8233 | |
|
| 0.9577 | 1.06 | 800 | 0.8399 | 0.7188 | 0.8387 | |
|
| 0.7999 | 1.2 | 900 | 0.7625 | 0.7396 | 0.8510 | |
|
| 0.7067 | 1.33 | 1000 | 0.7112 | 0.7534 | 0.8537 | |
|
| 0.6479 | 1.46 | 1100 | 0.6395 | 0.7807 | 0.8695 | |
|
| 0.6 | 1.6 | 1200 | 0.6111 | 0.8015 | 0.8781 | |
|
| 0.5168 | 1.73 | 1300 | 0.5787 | 0.8070 | 0.8783 | |
|
| 0.5635 | 1.86 | 1400 | 0.5333 | 0.8167 | 0.8873 | |
|
| 0.5094 | 1.99 | 1500 | 0.5283 | 0.8217 | 0.8868 | |
|
| 0.3862 | 2.13 | 1600 | 0.4973 | 0.8257 | 0.8908 | |
|
| 0.3663 | 2.26 | 1700 | 0.4879 | 0.8281 | 0.8889 | |
|
| 0.3584 | 2.39 | 1800 | 0.4619 | 0.8406 | 0.8973 | |
|
| 0.3427 | 2.53 | 1900 | 0.4460 | 0.8555 | 0.9044 | |
|
| 0.3334 | 2.66 | 2000 | 0.4386 | 0.8600 | 0.9056 | |
|
| 0.3267 | 2.79 | 2100 | 0.4274 | 0.8638 | 0.9064 | |
|
| 0.3045 | 2.93 | 2200 | 0.4154 | 0.8704 | 0.9094 | |
|
| 0.3048 | 3.06 | 2300 | 0.4156 | 0.8703 | 0.9106 | |
|
| 0.2329 | 3.19 | 2400 | 0.4068 | 0.8640 | 0.9097 | |
|
| 0.2393 | 3.32 | 2500 | 0.3957 | 0.8766 | 0.9122 | |
|
| 0.2335 | 3.46 | 2600 | 0.3923 | 0.8776 | 0.9159 | |
|
| 0.201 | 3.59 | 2700 | 0.3840 | 0.8810 | 0.9175 | |
|
| 0.2156 | 3.72 | 2800 | 0.3849 | 0.8817 | 0.9174 | |
|
| 0.2135 | 3.86 | 2900 | 0.3777 | 0.8833 | 0.9190 | |
|
| 0.2164 | 3.99 | 3000 | 0.3765 | 0.8829 | 0.9185 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.35.2 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.15.0 |
|
- Tokenizers 0.15.0 |
|
|
