hasoc19-microsoft-mdeberta-v3-base-HatredStatement-new
This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6039
- Accuracy: 0.7329
- Precision: 0.7324
- Recall: 0.7329
- F1: 0.7316
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 296 | 0.5276 | 0.7253 | 0.7258 | 0.7253 | 0.7225 |
| 0.5406 | 2.0 | 592 | 0.5513 | 0.7319 | 0.7348 | 0.7319 | 0.7278 |
| 0.5406 | 3.0 | 888 | 0.5466 | 0.7357 | 0.7458 | 0.7357 | 0.7283 |
| 0.4372 | 4.0 | 1184 | 0.5531 | 0.7452 | 0.7502 | 0.7452 | 0.7406 |
| 0.4372 | 5.0 | 1480 | 0.5927 | 0.7367 | 0.7364 | 0.7367 | 0.7352 |
| 0.3868 | 6.0 | 1776 | 0.6039 | 0.7329 | 0.7324 | 0.7329 | 0.7316 |
Framework versions
- Transformers 4.24.0.dev0
- Pytorch 1.11.0+cu102
- Datasets 2.6.1
- Tokenizers 0.13.1