bert-base-uncased-finetuned-cola_HW2_sepehr_bakhshi_dropout_00
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: 1.3418
- Matthews Correlation: 0.5910
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: 1.9628623388222396e-05
- train_batch_size: 16
- eval_batch_size: 16
- 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 |
---|---|---|---|---|
0.484 | 1.0 | 535 | 0.4557 | 0.5053 |
0.3013 | 2.0 | 1070 | 0.4224 | 0.5711 |
0.1949 | 3.0 | 1605 | 0.8633 | 0.5523 |
0.1399 | 4.0 | 2140 | 0.7826 | 0.5858 |
0.0933 | 5.0 | 2675 | 0.9575 | 0.5846 |
0.0607 | 6.0 | 3210 | 1.0032 | 0.5694 |
0.0554 | 7.0 | 3745 | 1.2276 | 0.5702 |
0.0368 | 8.0 | 4280 | 1.2437 | 0.5761 |
0.0303 | 9.0 | 4815 | 1.2978 | 0.5889 |
0.0146 | 10.0 | 5350 | 1.3418 | 0.5910 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
- Downloads last month
- 1
Dataset used to train sepehrbakhshi/bert-base-uncased-finetuned-cola_HW2_sepehr_bakhshi_dropout_00
Evaluation results
- Matthews Correlation on gluevalidation set self-reported0.591