<!-- 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. -->
bert_uncased_L-12_H-512_A-8_massive
This model is a fine-tuned version of google/bert_uncased_L-12_H-512_A-8 on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 0.5954
- Accuracy: 0.8874
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: 64
- eval_batch_size: 64
- seed: 33
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.4312 | 1.0 | 180 | 1.2711 | 0.7585 |
| 0.9941 | 2.0 | 360 | 0.7610 | 0.8372 |
| 0.5629 | 3.0 | 540 | 0.6013 | 0.8618 |
| 0.3608 | 4.0 | 720 | 0.5765 | 0.8711 |
| 0.2462 | 5.0 | 900 | 0.5568 | 0.8736 |
| 0.1717 | 6.0 | 1080 | 0.5698 | 0.8805 |
| 0.1212 | 7.0 | 1260 | 0.5840 | 0.8800 |
| 0.0877 | 8.0 | 1440 | 0.6068 | 0.8805 |
| 0.0629 | 9.0 | 1620 | 0.5954 | 0.8874 |
| 0.047 | 10.0 | 1800 | 0.6252 | 0.8829 |
| 0.0357 | 11.0 | 1980 | 0.6378 | 0.8834 |
| 0.0287 | 12.0 | 2160 | 0.6299 | 0.8854 |
| 0.0219 | 13.0 | 2340 | 0.6424 | 0.8849 |
| 0.0178 | 14.0 | 2520 | 0.6426 | 0.8849 |
| 0.0169 | 15.0 | 2700 | 0.6429 | 0.8854 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.1
