<!-- 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. -->
HBERTv1_48_L2_H512_A8_massive
This model is a fine-tuned version of gokuls/HBERTv1_48_L2_H512_A8 on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 0.6783
- Accuracy: 0.8618
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 |
|---|---|---|---|---|
| 2.0702 | 1.0 | 180 | 0.9458 | 0.7580 |
| 0.8247 | 2.0 | 360 | 0.6590 | 0.8234 |
| 0.5566 | 3.0 | 540 | 0.5933 | 0.8421 |
| 0.4096 | 4.0 | 720 | 0.5707 | 0.8529 |
| 0.3044 | 5.0 | 900 | 0.5893 | 0.8465 |
| 0.2319 | 6.0 | 1080 | 0.6023 | 0.8455 |
| 0.1782 | 7.0 | 1260 | 0.5828 | 0.8559 |
| 0.1356 | 8.0 | 1440 | 0.6171 | 0.8598 |
| 0.1048 | 9.0 | 1620 | 0.6322 | 0.8534 |
| 0.0806 | 10.0 | 1800 | 0.6544 | 0.8524 |
| 0.0648 | 11.0 | 1980 | 0.6718 | 0.8603 |
| 0.0516 | 12.0 | 2160 | 0.6737 | 0.8569 |
| 0.0383 | 13.0 | 2340 | 0.6783 | 0.8618 |
| 0.033 | 14.0 | 2520 | 0.7045 | 0.8534 |
| 0.0277 | 15.0 | 2700 | 0.7029 | 0.8564 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0
