<!-- 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_L8_H64_A2_emotion
This model is a fine-tuned version of gokuls/HBERTv1_48_L8_H64_A2 on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.5469
- Accuracy: 0.838
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 |
|---|---|---|---|---|
| 1.6018 | 1.0 | 250 | 1.4581 | 0.476 |
| 1.3077 | 2.0 | 500 | 1.1563 | 0.5335 |
| 1.0679 | 3.0 | 750 | 0.9798 | 0.6695 |
| 0.8802 | 4.0 | 1000 | 0.7861 | 0.7415 |
| 0.7364 | 5.0 | 1250 | 0.6777 | 0.804 |
| 0.6419 | 6.0 | 1500 | 0.6319 | 0.8145 |
| 0.5762 | 7.0 | 1750 | 0.5870 | 0.8295 |
| 0.5311 | 8.0 | 2000 | 0.5629 | 0.8345 |
| 0.5022 | 9.0 | 2250 | 0.5530 | 0.8355 |
| 0.4777 | 10.0 | 2500 | 0.5469 | 0.838 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.0
