<!-- 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_L12_H256_A4_emotion
This model is a fine-tuned version of gokuls/HBERTv1_48_L12_H256_A4 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.4661
- Accuracy: 0.868
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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.4452 | 1.0 | 250 | 1.2440 | 0.5355 |
| 1.1705 | 2.0 | 500 | 1.1210 | 0.5835 |
| 1.0848 | 3.0 | 750 | 1.0873 | 0.5925 |
| 0.9334 | 4.0 | 1000 | 0.8699 | 0.707 |
| 0.7142 | 5.0 | 1250 | 0.7357 | 0.7705 |
| 0.6285 | 6.0 | 1500 | 0.6807 | 0.795 |
| 0.518 | 7.0 | 1750 | 0.5330 | 0.8505 |
| 0.4191 | 8.0 | 2000 | 0.5028 | 0.859 |
| 0.3743 | 9.0 | 2250 | 0.4738 | 0.864 |
| 0.3383 | 10.0 | 2500 | 0.4661 | 0.868 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0
