<!-- 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. -->
DeBERTa_Offensive_English_Twitter
This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2527
- Macro F1: 0.9072
- Micro F1: 0.9114
- Accuracy: 0.9114
Results on the test set:
Accuracy: 0.9203707604803033
F1 score: 0.9030805039045081
Precision: 0.8948599346467434
Recall : 0.9127112237048041
Matthews Correlation Coefficient: 0.8073738336608549
Precision of each class: [0.95849624 0.83122363]
Recall of each class: [0.9299679 0.89545455]
F1 score of each class: [0.94401659 0.86214442]
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: 8
- eval_batch_size: 10
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Micro F1 | Accuracy |
|---|---|---|---|---|---|---|
| 0.3045 | 1.0 | 4036 | 0.3042 | 0.8903 | 0.8968 | 0.8968 |
| 0.2935 | 2.0 | 8073 | 0.3068 | 0.8877 | 0.8921 | 0.8921 |
| 0.2478 | 3.0 | 12109 | 0.2746 | 0.9031 | 0.9072 | 0.9072 |
| 0.2041 | 4.0 | 16144 | 0.2527 | 0.9072 | 0.9114 | 0.9114 |
Framework versions
- Transformers 4.27.1
- Pytorch 2.0.1+cu118
- Datasets 2.9.0
- Tokenizers 0.13.3
