bert-finetuned-mrpc
This model is a fine-tuned version of bert-base-cased on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.5152
- Accuracy: 0.8603
- F1: 0.9032
- Combined Score: 0.8818
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 16
- total_eval_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| No log | 1.0 | 230 | 0.3668 | 0.8431 | 0.8881 | 0.8656 |
| No log | 2.0 | 460 | 0.3751 | 0.8578 | 0.9017 | 0.8798 |
| 0.4264 | 3.0 | 690 | 0.5152 | 0.8603 | 0.9032 | 0.8818 |
Framework versions
- Transformers 4.11.0.dev0
- Pytorch 1.8.1+cu111
- Datasets 1.10.3.dev0
- Tokenizers 0.10.3