my_xlm-roberta-large-finetuned-conlljob04
This model is a fine-tuned version of xlm-roberta-large on the conll2003job dataset. It achieves the following results on the evaluation set:
- Loss: 0.0420
- Precision: 0.9617
- Recall: 0.9670
- F1: 0.9643
- Accuracy: 0.9936
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: 16
- eval_batch_size: 16
- 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 | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.1566 | 1.0 | 896 | 0.0403 | 0.9425 | 0.9542 | 0.9483 | 0.9911 |
| 0.0319 | 2.0 | 1792 | 0.0359 | 0.9523 | 0.9571 | 0.9547 | 0.9922 |
| 0.0156 | 3.0 | 2688 | 0.0356 | 0.9594 | 0.9625 | 0.9609 | 0.9929 |
| 0.01 | 4.0 | 3584 | 0.0377 | 0.9604 | 0.9672 | 0.9638 | 0.9934 |
| 0.0058 | 5.0 | 4480 | 0.0398 | 0.9618 | 0.9662 | 0.9640 | 0.9934 |
| 0.0034 | 6.0 | 5376 | 0.0420 | 0.9617 | 0.9670 | 0.9643 | 0.9936 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1