|
--- |
|
license: mit |
|
base_model: xlm-roberta-large |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: xlm-roberta-large_ALL_BCE_NEW_data_multihead_19_shuffled_special_tokens_val |
|
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. --> |
|
|
|
# xlm-roberta-large_ALL_BCE_NEW_data_multihead_19_shuffled_special_tokens_val |
|
|
|
This model is a fine-tuned version of [xlm-roberta-large](https://huggingface.co/xlm-roberta-large) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8445 |
|
- F1 Macro 0.1: 0.0895 |
|
- F1 Macro 0.15: 0.1160 |
|
- F1 Macro 0.2: 0.1402 |
|
- F1 Macro 0.25: 0.1634 |
|
- F1 Macro 0.3: 0.1847 |
|
- F1 Macro 0.35: 0.2040 |
|
- F1 Macro 0.4: 0.2229 |
|
- F1 Macro 0.45: 0.2406 |
|
- F1 Macro 0.5: 0.2583 |
|
- F1 Macro 0.55: 0.2763 |
|
- F1 Macro 0.6: 0.2924 |
|
- F1 Macro 0.65: 0.3101 |
|
- F1 Macro 0.7: 0.3251 |
|
- F1 Macro 0.75: 0.3405 |
|
- F1 Macro 0.8: 0.3547 |
|
- F1 Macro 0.85: 0.3634 |
|
- F1 Macro 0.9: 0.3572 |
|
- F1 Macro 0.95: 0.2839 |
|
- Threshold 0: 0.8 |
|
- Threshold 1: 0.85 |
|
- Threshold 2: 0.9 |
|
- Threshold 3: 0.9 |
|
- Threshold 4: 0.8 |
|
- Threshold 5: 0.85 |
|
- Threshold 6: 0.8 |
|
- Threshold 7: 0.9 |
|
- Threshold 8: 0.9 |
|
- Threshold 9: 0.8 |
|
- Threshold 10: 0.95 |
|
- Threshold 11: 0.85 |
|
- Threshold 12: 0.9 |
|
- Threshold 13: 0.8 |
|
- Threshold 14: 0.9 |
|
- Threshold 15: 0.85 |
|
- Threshold 16: 0.85 |
|
- Threshold 17: 0.85 |
|
- Threshold 18: 0.9 |
|
- 0: 0.1543 |
|
- 1: 0.2738 |
|
- 2: 0.3791 |
|
- 3: 0.2915 |
|
- 4: 0.4439 |
|
- 5: 0.4944 |
|
- 6: 0.4463 |
|
- 7: 0.3216 |
|
- 8: 0.3402 |
|
- 9: 0.5410 |
|
- 10: 0.5665 |
|
- 11: 0.5310 |
|
- 12: 0.2331 |
|
- 13: 0.1319 |
|
- 14: 0.3899 |
|
- 15: 0.3173 |
|
- 16: 0.4432 |
|
- 17: 0.6120 |
|
- 18: 0.2342 |
|
- Max F1: 0.3634 |
|
- Mean F1: 0.3761 |
|
|
|
## 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-06 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 2024 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- 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 | F1 Macro 0.1 | F1 Macro 0.15 | F1 Macro 0.2 | F1 Macro 0.25 | F1 Macro 0.3 | F1 Macro 0.35 | F1 Macro 0.4 | F1 Macro 0.45 | F1 Macro 0.5 | F1 Macro 0.55 | F1 Macro 0.6 | F1 Macro 0.65 | F1 Macro 0.7 | F1 Macro 0.75 | F1 Macro 0.8 | F1 Macro 0.85 | F1 Macro 0.9 | F1 Macro 0.95 | Threshold 0 | Threshold 1 | Threshold 2 | Threshold 3 | Threshold 4 | Threshold 5 | Threshold 6 | Threshold 7 | Threshold 8 | Threshold 9 | Threshold 10 | Threshold 11 | Threshold 12 | Threshold 13 | Threshold 14 | Threshold 15 | Threshold 16 | Threshold 17 | Threshold 18 | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | Max F1 | Mean F1 | |
|
|:-------------:|:-----:|:-----:|:---------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:-------:| |
|
| 1.2949 | 1.0 | 5595 | 0.9920 | 0.0638 | 0.0742 | 0.0860 | 0.0994 | 0.1129 | 0.1278 | 0.1430 | 0.1589 | 0.1751 | 0.1903 | 0.2064 | 0.2235 | 0.2373 | 0.2479 | 0.2512 | 0.2275 | 0.1775 | 0.0876 | 0.75 | 0.8 | 0.75 | 0.85 | 0.65 | 0.8 | 0.75 | 0.85 | 0.8 | 0.7 | 0.9 | 0.75 | 0.8 | 0.8 | 0.85 | 0.8 | 0.85 | 0.9 | 0.85 | 0.0863 | 0.1572 | 0.2169 | 0.0959 | 0.2903 | 0.3523 | 0.3723 | 0.1624 | 0.2313 | 0.4610 | 0.3852 | 0.4756 | 0.1678 | 0.1154 | 0.2816 | 0.1848 | 0.3673 | 0.5307 | 0.1168 | 0.2512 | 0.2658 | |
|
| 0.9147 | 2.0 | 11190 | 0.9023 | 0.0813 | 0.1044 | 0.1275 | 0.1498 | 0.1706 | 0.1898 | 0.2088 | 0.2261 | 0.2449 | 0.2624 | 0.2798 | 0.2951 | 0.3107 | 0.3233 | 0.3328 | 0.3348 | 0.3156 | 0.2286 | 0.75 | 0.8 | 0.85 | 0.9 | 0.75 | 0.85 | 0.8 | 0.85 | 0.8 | 0.8 | 0.9 | 0.85 | 0.9 | 0.65 | 0.9 | 0.9 | 0.85 | 0.9 | 0.95 | 0.1231 | 0.2517 | 0.3359 | 0.2514 | 0.4106 | 0.4565 | 0.4166 | 0.2556 | 0.3152 | 0.5241 | 0.5686 | 0.5085 | 0.2177 | 0.1176 | 0.3757 | 0.3059 | 0.4286 | 0.5881 | 0.2143 | 0.3348 | 0.3508 | |
|
| 0.732 | 3.0 | 16785 | 0.8445 | 0.0895 | 0.1160 | 0.1402 | 0.1634 | 0.1847 | 0.2040 | 0.2229 | 0.2406 | 0.2583 | 0.2763 | 0.2924 | 0.3101 | 0.3251 | 0.3405 | 0.3547 | 0.3634 | 0.3572 | 0.2839 | 0.8 | 0.85 | 0.9 | 0.9 | 0.8 | 0.85 | 0.8 | 0.9 | 0.9 | 0.8 | 0.95 | 0.85 | 0.9 | 0.8 | 0.9 | 0.85 | 0.85 | 0.85 | 0.9 | 0.1543 | 0.2738 | 0.3791 | 0.2915 | 0.4439 | 0.4944 | 0.4463 | 0.3216 | 0.3402 | 0.5410 | 0.5665 | 0.5310 | 0.2331 | 0.1319 | 0.3899 | 0.3173 | 0.4432 | 0.6120 | 0.2342 | 0.3634 | 0.3761 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.1 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.13.1 |
|
- Tokenizers 0.15.0 |
|
|