deberta-v3-large-go-emotions
This model is a fine-tuned version of microsoft/deberta-v3-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0811
- Macro F1: 0.4688
- Micro F1: 0.5960
- Weighted F1: 0.5778
- Macro Precision: 0.5670
- Macro Recall: 0.4250
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: 2e-05
- train_batch_size: 64
- eval_batch_size: 128
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Macro F1 | Micro F1 | Weighted F1 | Macro Precision | Macro Recall |
|---|---|---|---|---|---|---|---|---|
| 0.099 | 1.0 | 679 | 0.0897 | 0.3325 | 0.5469 | 0.4907 | 0.4971 | 0.2874 |
| 0.0814 | 2.0 | 1358 | 0.0825 | 0.4533 | 0.5843 | 0.5611 | 0.5792 | 0.3960 |
| 0.0694 | 3.0 | 2037 | 0.0826 | 0.4682 | 0.5986 | 0.5789 | 0.5499 | 0.4237 |
Framework versions
- Transformers 4.57.6
- Pytorch 2.4.1+cu124
- Datasets 4.8.4
- Tokenizers 0.22.2