metadata
library_name: transformers
license: apache-2.0
base_model: bert-base-multilingual-uncased
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: SA_Model_bert-base-multilingual-uncased
results: []
SA_Model_bert-base-multilingual-uncased
This model is a fine-tuned version of bert-base-multilingual-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6256
- Accuracy: 0.9000
- Precision: 0.8995
- Recall: 0.9000
- F1: 0.8998
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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| 0.4675 | 1.0 | 2004 | 0.3975 | 0.8631 | 0.8637 | 0.8631 | 0.8630 |
| 0.36 | 2.0 | 4008 | 0.3589 | 0.8695 | 0.8781 | 0.8695 | 0.8719 |
| 0.2715 | 3.0 | 6012 | 0.4023 | 0.8822 | 0.8840 | 0.8822 | 0.8828 |
| 0.222 | 4.0 | 8016 | 0.4390 | 0.8917 | 0.8925 | 0.8917 | 0.8920 |
| 0.151 | 5.0 | 10020 | 0.5550 | 0.8994 | 0.8984 | 0.8994 | 0.8981 |
| 0.1544 | 6.0 | 12024 | 0.6256 | 0.9000 | 0.8995 | 0.9000 | 0.8998 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.12.0+cu130
- Datasets 3.6.0
- Tokenizers 0.21.1