|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: microsoft/deberta-v3-base |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- accuracy |
|
|
- precision |
|
|
- recall |
|
|
- f1 |
|
|
model-index: |
|
|
- name: deberta-base |
|
|
results: [] |
|
|
--- |
|
|
|
|
|
<!-- 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. --> |
|
|
|
|
|
# deberta-base |
|
|
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.1665 |
|
|
- Accuracy: 0.9601 |
|
|
- Precision: 0.9599 |
|
|
- Recall: 0.9601 |
|
|
- F1: 0.9594 |
|
|
- Auroc: 0.9928 |
|
|
|
|
|
## 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: 16 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 64 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 1 |
|
|
- label_smoothing_factor: 0.03 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | Auroc | |
|
|
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:------:| |
|
|
| 0.4866 | 0.0988 | 256 | 0.2931 | 0.8845 | 0.8939 | 0.8845 | 0.8876 | 0.9465 | |
|
|
| 0.2757 | 0.1977 | 512 | 0.3478 | 0.8898 | 0.8984 | 0.8898 | 0.8765 | 0.9544 | |
|
|
| 0.2433 | 0.2965 | 768 | 0.2097 | 0.9404 | 0.9413 | 0.9404 | 0.9408 | 0.9799 | |
|
|
| 0.2332 | 0.3953 | 1024 | 0.3548 | 0.8815 | 0.8907 | 0.8815 | 0.8657 | 0.9690 | |
|
|
| 0.2152 | 0.4942 | 1280 | 0.1942 | 0.9440 | 0.9434 | 0.9440 | 0.9426 | 0.9868 | |
|
|
| 0.1907 | 0.5930 | 1536 | 0.1615 | 0.9649 | 0.9647 | 0.9649 | 0.9647 | 0.9899 | |
|
|
| 0.1865 | 0.6918 | 1792 | 0.1556 | 0.9655 | 0.9654 | 0.9655 | 0.9654 | 0.9922 | |
|
|
| 0.1865 | 0.7907 | 2048 | 0.2322 | 0.9369 | 0.9370 | 0.9369 | 0.9344 | 0.9773 | |
|
|
| 0.168 | 0.8895 | 2304 | 0.1653 | 0.9672 | 0.9670 | 0.9672 | 0.9668 | 0.9937 | |
|
|
| 0.1732 | 0.9883 | 2560 | 0.1467 | 0.9702 | 0.9716 | 0.9702 | 0.9706 | 0.9935 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.45.1 |
|
|
- Pytorch 2.4.0 |
|
|
- Datasets 3.0.1 |
|
|
- Tokenizers 0.20.0 |
|
|
|
