hasoc19-xlm-roberta-base-sentiment-new
This model is a fine-tuned version of xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3840
- Accuracy: 0.8726
- Precision: 0.8724
- Recall: 0.8726
- F1: 0.8725
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 |
|---|---|---|---|---|---|---|---|
| 0.4786 | 1.0 | 537 | 0.3999 | 0.8381 | 0.8391 | 0.8381 | 0.8363 |
| 0.349 | 2.0 | 1074 | 0.3443 | 0.8606 | 0.8603 | 0.8606 | 0.8603 |
| 0.2927 | 3.0 | 1611 | 0.3412 | 0.8669 | 0.8668 | 0.8669 | 0.8662 |
| 0.2471 | 4.0 | 2148 | 0.3408 | 0.8705 | 0.8708 | 0.8705 | 0.8706 |
| 0.2195 | 5.0 | 2685 | 0.3897 | 0.8726 | 0.8725 | 0.8726 | 0.8721 |
| 0.1849 | 6.0 | 3222 | 0.3840 | 0.8726 | 0.8724 | 0.8726 | 0.8725 |
Framework versions
- Transformers 4.24.0.dev0
- Pytorch 1.11.0+cu102
- Datasets 2.6.1
- Tokenizers 0.13.1