bert_cyber
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5817
- Accuracy: 0.8091
- F1: 0.7970
- Precision: 0.7905
- Recall: 0.8138
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
---|---|---|---|---|---|---|---|
0.4515 | 1.0 | 144 | 0.4555 | 0.7778 | 0.7723 | 0.7790 | 0.8103 |
0.3411 | 2.0 | 288 | 0.3886 | 0.8143 | 0.7992 | 0.7933 | 0.8089 |
0.2876 | 3.0 | 432 | 0.4959 | 0.7893 | 0.7826 | 0.7845 | 0.8161 |
0.2637 | 4.0 | 576 | 0.5285 | 0.7950 | 0.7882 | 0.7889 | 0.8208 |
0.1519 | 5.0 | 720 | 0.5416 | 0.8127 | 0.8001 | 0.7934 | 0.8151 |
0.1076 | 6.0 | 864 | 0.5817 | 0.8091 | 0.7970 | 0.7905 | 0.8138 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 2