metadata
license: mit
base_model: microsoft/deberta-v3-large
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: deberta-v3-large-zeroshot-v2.0-2024-04-01-09-59
results: []
deberta-v3-large-zeroshot-v2.0-2024-04-01-09-59
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: 0.1590
- F1 Macro: 0.6865
- F1 Micro: 0.7525
- Accuracy Balanced: 0.7303
- Accuracy: 0.7525
- Precision Macro: 0.6989
- Recall Macro: 0.7303
- Precision Micro: 0.7525
- Recall Micro: 0.7525
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: 9e-06
- train_batch_size: 4
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Micro | Accuracy Balanced | Accuracy | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.1929 | 1.0 | 43383 | 0.3181 | 0.8694 | 0.8813 | 0.8688 | 0.8813 | 0.8701 | 0.8688 | 0.8813 | 0.8813 |
| 0.1358 | 2.0 | 86766 | 0.3519 | 0.8730 | 0.8849 | 0.8713 | 0.8849 | 0.8749 | 0.8713 | 0.8849 | 0.8849 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.1.2+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2