metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: roberta-base-academic3
results: []
roberta-base-academic3
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.4206
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: 7e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 64
- total_train_batch_size: 512
- optimizer: Adam with betas=(0.9,0.99) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.2
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.6943 | 0.99 | 82 | 1.5540 |
| 1.6494 | 1.99 | 164 | 1.5268 |
| 1.63 | 2.99 | 246 | 1.5209 |
| 1.6152 | 3.99 | 328 | 1.5049 |
| 1.5985 | 4.99 | 410 | 1.4891 |
| 1.5826 | 5.99 | 492 | 1.4876 |
| 1.5643 | 6.99 | 574 | 1.4769 |
| 1.5506 | 7.99 | 656 | 1.4638 |
| 1.5383 | 8.99 | 738 | 1.4548 |
| 1.5309 | 9.99 | 820 | 1.4511 |
| 1.5225 | 10.99 | 902 | 1.4492 |
| 1.5124 | 11.99 | 984 | 1.4419 |
| 1.507 | 12.99 | 1066 | 1.4323 |
| 1.4985 | 13.99 | 1148 | 1.4294 |
| 1.4921 | 14.99 | 1230 | 1.4296 |
| 1.4859 | 15.99 | 1312 | 1.4256 |
| 1.4827 | 16.99 | 1394 | 1.4194 |
| 1.4756 | 17.99 | 1476 | 1.4184 |
| 1.474 | 18.99 | 1558 | 1.4156 |
| 1.4737 | 19.99 | 1640 | 1.4165 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1+cu116
- Datasets 2.8.0
- Tokenizers 0.13.2