deberta-v3-xsmall-emotion
This model is a fine-tuned version of microsoft/deberta-v3-xsmall on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.1877
- Accuracy: 0.932
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.3683 | 1.0 | 500 | 0.8479 | 0.6975 |
| 0.547 | 2.0 | 1000 | 0.2881 | 0.905 |
| 0.2378 | 3.0 | 1500 | 0.2116 | 0.925 |
| 0.1704 | 4.0 | 2000 | 0.1877 | 0.932 |
| 0.1392 | 5.0 | 2500 | 0.1718 | 0.9295 |
Framework versions
- Transformers 4.12.3
- Pytorch 1.9.1
- Datasets 1.15.1
- Tokenizers 0.10.3