<!-- 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. -->
hbertv2-wt-frz-48-Massive-intent
This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48_frz on the massive dataset. It achieves the following results on the evaluation set:
- Loss: 0.8549
- Accuracy: 0.8701
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 |
|---|---|---|---|---|
| 1.6771 | 1.0 | 180 | 0.8299 | 0.7850 |
| 0.7544 | 2.0 | 360 | 0.6817 | 0.8185 |
| 0.5364 | 3.0 | 540 | 0.6402 | 0.8362 |
| 0.4001 | 4.0 | 720 | 0.6371 | 0.8367 |
| 0.2985 | 5.0 | 900 | 0.6864 | 0.8367 |
| 0.2297 | 6.0 | 1080 | 0.6357 | 0.8485 |
| 0.1633 | 7.0 | 1260 | 0.7224 | 0.8411 |
| 0.1304 | 8.0 | 1440 | 0.7212 | 0.8593 |
| 0.0859 | 9.0 | 1620 | 0.7789 | 0.8515 |
| 0.0632 | 10.0 | 1800 | 0.8223 | 0.8588 |
| 0.0447 | 11.0 | 1980 | 0.8011 | 0.8628 |
| 0.0288 | 12.0 | 2160 | 0.8139 | 0.8692 |
| 0.0188 | 13.0 | 2340 | 0.8859 | 0.8662 |
| 0.0115 | 14.0 | 2520 | 0.8549 | 0.8701 |
| 0.0067 | 15.0 | 2700 | 0.8622 | 0.8677 |
Framework versions
- Transformers 4.31.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.1
- Tokenizers 0.13.3
