--- 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 results: [] --- # xlm-roberta-large_ALL_BCE_new_data_multihead_19_shuffled_special_tokens 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