distilbert_add_GLUE_Experiment_qqp_384
This model is a fine-tuned version of distilbert-base-uncased on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.4096
- Accuracy: 0.8095
- F1: 0.7372
- Combined Score: 0.7734
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: 256
- eval_batch_size: 256
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.5518 | 1.0 | 1422 | 0.5289 | 0.7376 | 0.6535 | 0.6955 |
| 0.4901 | 2.0 | 2844 | 0.4655 | 0.7772 | 0.6744 | 0.7258 |
| 0.4098 | 3.0 | 4266 | 0.4096 | 0.8095 | 0.7372 | 0.7734 |
| 0.3273 | 4.0 | 5688 | 0.4343 | 0.8211 | 0.7536 | 0.7873 |
| 0.2681 | 5.0 | 7110 | 0.4322 | 0.8286 | 0.7519 | 0.7902 |
| 0.223 | 6.0 | 8532 | 0.4789 | 0.8301 | 0.7502 | 0.7901 |
| 0.1883 | 7.0 | 9954 | 0.4715 | 0.8329 | 0.7663 | 0.7996 |
| 0.1603 | 8.0 | 11376 | 0.5090 | 0.8346 | 0.7577 | 0.7961 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2