<!-- 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. -->
multiclass_hate_speech_detection
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: 0.4805
- Macro F1: 0.8349
- Micro F1: 0.8598
- Accuracy: 0.8598
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-06
- train_batch_size: 10
- eval_batch_size: 10
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 20
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Micro F1 | Accuracy |
|---|---|---|---|---|---|---|
| 0.4573 | 1.0 | 2522 | 0.4370 | 0.7906 | 0.8243 | 0.8243 |
| 0.3809 | 2.0 | 5044 | 0.3856 | 0.8146 | 0.8408 | 0.8408 |
| 0.3221 | 3.0 | 7566 | 0.3914 | 0.8191 | 0.8451 | 0.8451 |
| 0.2888 | 4.0 | 10088 | 0.3837 | 0.8262 | 0.8502 | 0.8502 |
| 0.2492 | 5.0 | 12610 | 0.3876 | 0.8298 | 0.8540 | 0.8540 |
| 0.2305 | 6.0 | 15132 | 0.4109 | 0.8282 | 0.8532 | 0.8532 |
| 0.213 | 7.0 | 17654 | 0.4296 | 0.8315 | 0.8574 | 0.8574 |
| 0.1839 | 8.0 | 20176 | 0.4423 | 0.8348 | 0.8605 | 0.8605 |
| 0.1736 | 9.0 | 22698 | 0.4788 | 0.8336 | 0.8586 | 0.8586 |
| 0.154 | 10.0 | 25220 | 0.4805 | 0.8349 | 0.8598 | 0.8598 |
Framework versions
- Transformers 4.27.1
- Pytorch 2.0.1+cu118
- Datasets 2.9.0
- Tokenizers 0.13.3
