roberta-base-finetuned-cola
This model is a fine-tuned version of roberta-base on the glue dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5732
- Matthews Correlation: 0.6495
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: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: IPU
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- total_eval_batch_size: 5
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- training precision: Mixed Precision
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Matthews Correlation |
| 0.5211 |
1.0 |
534 |
0.4031 |
0.5599 |
| 0.3739 |
2.0 |
1068 |
0.4688 |
0.5713 |
| 0.0697 |
3.0 |
1602 |
0.4988 |
0.6070 |
| 0.0712 |
4.0 |
2136 |
0.5596 |
0.6221 |
| 0.0955 |
5.0 |
2670 |
0.5732 |
0.6495 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.10.0+cpu
- Datasets 2.8.0
- Tokenizers 0.12.1
