CNEC1_1_62types_xlm-roberta-large
This model is a fine-tuned version of FacebookAI/xlm-roberta-large on the cnec dataset. It achieves the following results on the evaluation set:
- Loss: 0.2611
- Precision: 0.8268
- Recall: 0.8836
- F1: 0.8543
- Accuracy: 0.9555
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
- lr_scheduler_warmup_ratio: 0.1
- lr_scheduler_warmup_steps: 1000
- num_epochs: 15
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
1.601 | 1.7 | 500 | 0.4369 | 0.6449 | 0.6810 | 0.6625 | 0.9094 |
0.3697 | 3.4 | 1000 | 0.2542 | 0.7377 | 0.8176 | 0.7756 | 0.9440 |
0.2295 | 5.1 | 1500 | 0.2484 | 0.7583 | 0.8523 | 0.8026 | 0.9449 |
0.1531 | 6.8 | 2000 | 0.2301 | 0.7952 | 0.8710 | 0.8314 | 0.9496 |
0.1107 | 8.5 | 2500 | 0.2284 | 0.8192 | 0.8729 | 0.8452 | 0.9529 |
0.0801 | 10.2 | 3000 | 0.2435 | 0.8308 | 0.8901 | 0.8595 | 0.9561 |
0.0621 | 11.9 | 3500 | 0.2507 | 0.8156 | 0.8794 | 0.8463 | 0.9544 |
0.0454 | 13.61 | 4000 | 0.2611 | 0.8268 | 0.8836 | 0.8543 | 0.9555 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 12
Finetuned from
Evaluation results
- Precision on cnecvalidation set self-reported0.827
- Recall on cnecvalidation set self-reported0.884
- F1 on cnecvalidation set self-reported0.854
- Accuracy on cnecvalidation set self-reported0.956