metadata
library_name: transformers
license: mit
base_model: FacebookAI/roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: roberta-base-pr_tqacd
results: []
roberta-base-pr_tqacd
This model is a fine-tuned version of FacebookAI/roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6056
- F1 Macro: 0.5289
- Precision: 0.5397
- Recall: 0.5373
- Accuracy: 0.7180
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: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | Precision | Recall | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 354 | 1.8365 | 0.3820 | 0.4312 | 0.4000 | 0.6106 |
| 2.2658 | 2.0 | 708 | 1.3285 | 0.5013 | 0.5037 | 0.5722 | 0.6554 |
| 1.5226 | 3.0 | 1062 | 1.2491 | 0.5450 | 0.5421 | 0.6006 | 0.7090 |
| 1.5226 | 4.0 | 1416 | 1.3314 | 0.5476 | 0.5555 | 0.5767 | 0.7180 |
| 1.1009 | 5.0 | 1770 | 1.4114 | 0.5468 | 0.5457 | 0.5754 | 0.7203 |
| 0.7195 | 6.0 | 2124 | 1.6056 | 0.5289 | 0.5397 | 0.5373 | 0.7180 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.8.0+cu128
- Datasets 4.4.1
- Tokenizers 0.22.1