|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: microsoft/mdeberta-v3-base |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: mdeberta-semeval25_narratives_fold5 |
|
|
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. --> |
|
|
|
|
|
# mdeberta-semeval25_narratives_fold5 |
|
|
|
|
|
This model is a fine-tuned version of [microsoft/mdeberta-v3-base](https://huggingface.co/microsoft/mdeberta-v3-base) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 4.0196 |
|
|
- Precision Samples: 0.3300 |
|
|
- Recall Samples: 0.8054 |
|
|
- F1 Samples: 0.4401 |
|
|
- Precision Macro: 0.6768 |
|
|
- Recall Macro: 0.5901 |
|
|
- F1 Macro: 0.3682 |
|
|
- Precision Micro: 0.2997 |
|
|
- Recall Micro: 0.7707 |
|
|
- F1 Micro: 0.4316 |
|
|
- Precision Weighted: 0.4440 |
|
|
- Recall Weighted: 0.7707 |
|
|
- F1 Weighted: 0.3890 |
|
|
|
|
|
## 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: 32 |
|
|
- eval_batch_size: 32 |
|
|
- 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 |
|
|
- num_epochs: 10 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision Samples | Recall Samples | F1 Samples | Precision Macro | Recall Macro | F1 Macro | Precision Micro | Recall Micro | F1 Micro | Precision Weighted | Recall Weighted | F1 Weighted | |
|
|
|:-------------:|:-----:|:----:|:---------------:|:-----------------:|:--------------:|:----------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:--------:|:------------------:|:---------------:|:-----------:| |
|
|
| 5.5606 | 1.0 | 19 | 5.1746 | 0.5241 | 0.0760 | 0.0945 | 0.9639 | 0.1737 | 0.1596 | 0.2418 | 0.0827 | 0.1232 | 0.9031 | 0.0827 | 0.0450 | |
|
|
| 4.8516 | 2.0 | 38 | 4.9279 | 0.2575 | 0.4866 | 0.3149 | 0.8592 | 0.3182 | 0.2199 | 0.2573 | 0.4323 | 0.3226 | 0.6527 | 0.4323 | 0.1902 | |
|
|
| 5.1102 | 3.0 | 57 | 4.6329 | 0.3105 | 0.6392 | 0.3944 | 0.7662 | 0.4137 | 0.2809 | 0.3010 | 0.5827 | 0.3969 | 0.5232 | 0.5827 | 0.2907 | |
|
|
| 4.5152 | 4.0 | 76 | 4.4162 | 0.2982 | 0.7060 | 0.3926 | 0.7663 | 0.4692 | 0.2933 | 0.2827 | 0.6654 | 0.3969 | 0.5256 | 0.6654 | 0.3110 | |
|
|
| 4.3922 | 5.0 | 95 | 4.2955 | 0.3114 | 0.7290 | 0.4139 | 0.7003 | 0.5100 | 0.3321 | 0.2961 | 0.6880 | 0.4140 | 0.4569 | 0.6880 | 0.3560 | |
|
|
| 4.0885 | 6.0 | 114 | 4.1427 | 0.3210 | 0.8169 | 0.4335 | 0.6788 | 0.5895 | 0.3665 | 0.2921 | 0.7820 | 0.4254 | 0.4415 | 0.7820 | 0.3845 | |
|
|
| 3.9996 | 7.0 | 133 | 4.0937 | 0.3164 | 0.7928 | 0.4286 | 0.6762 | 0.5803 | 0.3656 | 0.2945 | 0.7594 | 0.4244 | 0.4386 | 0.7594 | 0.3814 | |
|
|
| 3.9713 | 8.0 | 152 | 4.0603 | 0.3159 | 0.7847 | 0.4253 | 0.6727 | 0.5768 | 0.3623 | 0.2935 | 0.7481 | 0.4216 | 0.4375 | 0.7481 | 0.3792 | |
|
|
| 4.016 | 9.0 | 171 | 4.0393 | 0.3189 | 0.7905 | 0.4300 | 0.6750 | 0.5812 | 0.3654 | 0.2978 | 0.7556 | 0.4272 | 0.4418 | 0.7556 | 0.3848 | |
|
|
| 3.9635 | 10.0 | 190 | 4.0196 | 0.3300 | 0.8054 | 0.4401 | 0.6768 | 0.5901 | 0.3682 | 0.2997 | 0.7707 | 0.4316 | 0.4440 | 0.7707 | 0.3890 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.46.0 |
|
|
- Pytorch 2.3.1 |
|
|
- Datasets 2.21.0 |
|
|
- Tokenizers 0.20.1 |
|
|
|
