bart-base-finetuned-mrpc
This model is a fine-tuned version of facebook/bart-base on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.3674
- F1: 0.9062
- Accuracy: 0.8676
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: 32
- eval_batch_size: 32
- seed: 45
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 | Accuracy |
|---|---|---|---|---|---|
| No log | 1.0 | 115 | 0.4762 | 0.8663 | 0.7966 |
| No log | 2.0 | 230 | 0.3451 | 0.9019 | 0.8603 |
| No log | 3.0 | 345 | 0.3229 | 0.9028 | 0.8627 |
| No log | 4.0 | 460 | 0.3236 | 0.9014 | 0.8627 |
| 0.3737 | 5.0 | 575 | 0.3674 | 0.9062 | 0.8676 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3