metadata
license: apache-2.0
base_model: google-bert/bert-base-multilingual-cased
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: cyber_deberta
results: []
cyber_deberta
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: 0.4669
- Accuracy: 0.8315
- F1: 0.8135
- Precision: 0.8121
- Recall: 0.8150
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: 1e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.5788 | 1.0 | 105 | 0.5623 | 0.6755 | 0.4813 | 0.6766 | 0.5352 |
| 0.478 | 2.0 | 210 | 0.4430 | 0.7746 | 0.7444 | 0.7501 | 0.7401 |
| 0.4087 | 3.0 | 315 | 0.3948 | 0.8096 | 0.7835 | 0.7911 | 0.7777 |
| 0.4004 | 4.0 | 420 | 0.3868 | 0.8080 | 0.7917 | 0.7864 | 0.7998 |
| 0.3216 | 5.0 | 525 | 0.4005 | 0.8106 | 0.7928 | 0.7888 | 0.7980 |
| 0.3144 | 6.0 | 630 | 0.3878 | 0.8299 | 0.8062 | 0.8153 | 0.7994 |
| 0.2598 | 7.0 | 735 | 0.4040 | 0.8258 | 0.8084 | 0.8053 | 0.8121 |
| 0.2234 | 8.0 | 840 | 0.4280 | 0.8284 | 0.8108 | 0.8083 | 0.8137 |
| 0.2088 | 9.0 | 945 | 0.4580 | 0.8320 | 0.8154 | 0.8121 | 0.8194 |
| 0.1775 | 10.0 | 1050 | 0.4669 | 0.8315 | 0.8135 | 0.8121 | 0.8150 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1