bert-base-uncased-finetuned-cola-batch-4
This model is a fine-tuned version of bert-base-uncased on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 1.3628
- Matthews Correlation: 0.5990
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
---|---|---|---|---|
0.5495 | 1.0 | 2138 | 0.7520 | 0.4570 |
0.457 | 2.0 | 4276 | 0.8038 | 0.5567 |
0.2524 | 3.0 | 6414 | 0.9339 | 0.5416 |
0.1602 | 4.0 | 8552 | 1.0277 | 0.5809 |
0.1241 | 5.0 | 10690 | 1.2164 | 0.5830 |
0.1057 | 6.0 | 12828 | 1.2966 | 0.5855 |
0.0428 | 7.0 | 14966 | 1.3628 | 0.5990 |
0.0311 | 8.0 | 17104 | 1.3782 | 0.5843 |
0.0281 | 9.0 | 19242 | 1.6510 | 0.5452 |
0.0067 | 10.0 | 21380 | 1.5954 | 0.5713 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 13
Dataset used to train cansurav/bert-base-uncased-finetuned-cola-batch-4
Evaluation results
- Matthews Correlation on gluevalidation set self-reported0.599