metadata
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model_index:
- name: portuguese-archival-finding-aids
results:
- task:
name: Token Classification
type: token-classification
metric:
name: Accuracy
type: accuracy
value: 0.9617770479839446
portuguese-archival-finding-aids
This model is a fine-tuned version of bert-base-multilingual-cased on an unkown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1812
- Precision: 0.8624
- Recall: 0.9557
- F1: 0.9067
- Accuracy: 0.9618
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 192 | 0.1565 | 0.8511 | 0.9327 | 0.8900 | 0.9563 |
| 0.1849 | 2.0 | 384 | 0.1594 | 0.8634 | 0.9543 | 0.9065 | 0.9619 |
| 0.0454 | 3.0 | 576 | 0.1812 | 0.8624 | 0.9557 | 0.9067 | 0.9618 |
Framework versions
- Transformers 4.10.0.dev0
- Pytorch 1.9.0+cu111
- Datasets 1.10.2
- Tokenizers 0.10.3