BertAbsIntroComp
This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4389
- Accuracy: 0.8741
- Precision: 0.7956
- Recall: 0.7974
- F1: 0.7934
- Top3: 0.9775
- Top3macro: 0.9532
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: 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: 4
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Top3 | Top3macro |
---|---|---|---|---|---|---|---|---|---|
0.5767 | 1.0 | 7566 | 0.5428 | 0.8329 | 0.7224 | 0.6959 | 0.7029 | 0.9608 | 0.9153 |
0.3891 | 2.0 | 15132 | 0.4542 | 0.8600 | 0.7741 | 0.7570 | 0.7611 | 0.9748 | 0.9436 |
0.2962 | 3.0 | 22698 | 0.4540 | 0.8699 | 0.7881 | 0.7918 | 0.7884 | 0.9763 | 0.9499 |
0.1971 | 4.0 | 30264 | 0.4888 | 0.8736 | 0.7993 | 0.7941 | 0.7957 | 0.9785 | 0.9555 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.2
- Datasets 2.19.2
- Tokenizers 0.19.1
- Downloads last month
- 2