<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert_sa_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: 0.4322
- Accuracy: 0.8082
- F1: 0.7405
- Combined Score: 0.7744
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 |
---|---|---|---|---|---|---|
0.5251 | 1.0 | 1422 | 0.5016 | 0.7563 | 0.6686 | 0.7124 |
0.466 | 2.0 | 2844 | 0.4668 | 0.7745 | 0.6459 | 0.7102 |
0.4292 | 3.0 | 4266 | 0.4609 | 0.7854 | 0.6685 | 0.7270 |
0.3971 | 4.0 | 5688 | 0.4463 | 0.7945 | 0.7190 | 0.7568 |
0.3677 | 5.0 | 7110 | 0.4326 | 0.8001 | 0.7280 | 0.7641 |
0.3398 | 6.0 | 8532 | 0.4511 | 0.8017 | 0.7361 | 0.7689 |
0.3141 | 7.0 | 9954 | 0.4322 | 0.8082 | 0.7405 | 0.7744 |
0.2891 | 8.0 | 11376 | 0.4373 | 0.8096 | 0.7434 | 0.7765 |
0.266 | 9.0 | 12798 | 0.4793 | 0.7966 | 0.7440 | 0.7703 |
0.2433 | 10.0 | 14220 | 0.5018 | 0.8143 | 0.7503 | 0.7823 |
0.2235 | 11.0 | 15642 | 0.4917 | 0.8144 | 0.7428 | 0.7786 |
0.2045 | 12.0 | 17064 | 0.5152 | 0.8166 | 0.7521 | 0.7844 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2