<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
roberta-base-mnli_VALUE
This model is a fine-tuned version of WillHeld/roberta-base-mnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5880
- Acc: 0.8680
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: 64
- eval_batch_size: 64
- seed: 42
- 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 | Acc |
|---|---|---|---|---|
| 0.3487 | 0.33 | 2000 | 0.3948 | 0.8588 |
| 0.3008 | 0.65 | 4000 | 0.4038 | 0.8605 |
| 0.2767 | 0.98 | 6000 | 0.3889 | 0.8648 |
| 0.2186 | 1.3 | 8000 | 0.4352 | 0.8654 |
| 0.2177 | 1.63 | 10000 | 0.4251 | 0.8654 |
| 0.2212 | 1.96 | 12000 | 0.4155 | 0.8681 |
| 0.1638 | 2.28 | 14000 | 0.4783 | 0.8638 |
| 0.1592 | 2.61 | 16000 | 0.4572 | 0.8673 |
| 0.1565 | 2.93 | 18000 | 0.4669 | 0.8683 |
| 0.1213 | 3.26 | 20000 | 0.5114 | 0.8693 |
| 0.1154 | 3.59 | 22000 | 0.5604 | 0.8668 |
| 0.1159 | 3.91 | 24000 | 0.5143 | 0.8707 |
| 0.0952 | 4.24 | 26000 | 0.5900 | 0.8661 |
| 0.0881 | 4.56 | 28000 | 0.6000 | 0.8663 |
| 0.0879 | 4.89 | 30000 | 0.5880 | 0.8680 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.12.1
