metadata
library_name: transformers
license: mit
base_model: FacebookAI/roberta-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- accuracy
model-index:
- name: roberta-base-tqacd
results: []
roberta-base-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: 2.7467
- F1 Macro: 0.2528
- Precision: 0.2683
- Recall: 0.2648
- Accuracy: 0.3515
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 | 114 | 2.3961 | 0.0436 | 0.0603 | 0.1103 | 0.1584 |
| No log | 2.0 | 228 | 2.3381 | 0.1234 | 0.1119 | 0.1476 | 0.3564 |
| No log | 3.0 | 342 | 2.1873 | 0.2052 | 0.3773 | 0.2457 | 0.2673 |
| No log | 4.0 | 456 | 2.2524 | 0.1800 | 0.2231 | 0.2043 | 0.3267 |
| 2.241 | 5.0 | 570 | 2.2141 | 0.2128 | 0.2340 | 0.2494 | 0.3218 |
| 2.241 | 6.0 | 684 | 2.3365 | 0.2238 | 0.2480 | 0.2391 | 0.2822 |
| 2.241 | 7.0 | 798 | 2.4779 | 0.2805 | 0.3501 | 0.2756 | 0.3713 |
| 2.241 | 8.0 | 912 | 2.5194 | 0.2518 | 0.2908 | 0.2667 | 0.3416 |
| 0.9793 | 9.0 | 1026 | 2.7467 | 0.2528 | 0.2683 | 0.2648 | 0.3515 |
Framework versions
- Transformers 4.57.1
- Pytorch 2.8.0+cu128
- Datasets 4.4.1
- Tokenizers 0.22.1