RoBERTa_conll_learning_rate2e5
This model is a fine-tuned version of distilroberta-base on the conll2003 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0558
- Precision: 0.9394
- Recall: 0.9527
- F1: 0.9460
- Accuracy: 0.9877
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.0747 | 1.0 | 1756 | 0.0614 | 0.9147 | 0.9382 | 0.9263 | 0.9838 |
0.0426 | 2.0 | 3512 | 0.0558 | 0.9354 | 0.9498 | 0.9426 | 0.9870 |
0.0296 | 3.0 | 5268 | 0.0558 | 0.9394 | 0.9527 | 0.9460 | 0.9877 |
Framework versions
- Transformers 4.40.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 0
Finetuned from
Dataset used to train ICT2214Team7/RoBERTa_conll_learning_rate2e5
Evaluation results
- Precision on conll2003validation set self-reported0.939
- Recall on conll2003validation set self-reported0.953
- F1 on conll2003validation set self-reported0.946
- Accuracy on conll2003validation set self-reported0.988