--- license: mit base_model: roberta-large tags: - generated_from_trainer model-index: - name: roberta-large_ALL_BCE_translations_multihead_19_shuffled_special_tokens results: [] --- # roberta-large_ALL_BCE_translations_multihead_19_shuffled_special_tokens This model is a fine-tuned version of [roberta-large](https://huggingface.co/roberta-large) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.8569 - F1 Macro 0.1: 0.1078 - F1 Macro 0.15: 0.1410 - F1 Macro 0.2: 0.1700 - F1 Macro 0.25: 0.1957 - F1 Macro 0.3: 0.2179 - F1 Macro 0.35: 0.2383 - F1 Macro 0.4: 0.2580 - F1 Macro 0.45: 0.2757 - F1 Macro 0.5: 0.2935 - F1 Macro 0.55: 0.3110 - F1 Macro 0.6: 0.3275 - F1 Macro 0.65: 0.3425 - F1 Macro 0.7: 0.3592 - F1 Macro 0.75: 0.3717 - F1 Macro 0.8: 0.3829 - F1 Macro 0.85: 0.3903 - F1 Macro 0.9: 0.3847 - F1 Macro 0.95: 0.3225 - Threshold 0: 0.85 - Threshold 1: 0.8 - Threshold 2: 0.9 - Threshold 3: 0.9 - Threshold 4: 0.8 - Threshold 5: 0.8 - Threshold 6: 0.8 - Threshold 7: 0.9 - Threshold 8: 0.85 - Threshold 9: 0.8 - Threshold 10: 0.9 - Threshold 11: 0.85 - Threshold 12: 0.9 - Threshold 13: 0.85 - Threshold 14: 0.85 - Threshold 15: 0.9 - Threshold 16: 0.85 - Threshold 17: 0.9 - Threshold 18: 0.9 - 0: 0.1654 - 1: 0.3112 - 2: 0.3764 - 3: 0.3436 - 4: 0.4800 - 5: 0.4880 - 6: 0.4593 - 7: 0.3694 - 8: 0.3882 - 9: 0.5533 - 10: 0.5439 - 11: 0.5492 - 12: 0.2443 - 13: 0.2278 - 14: 0.4014 - 15: 0.3373 - 16: 0.4511 - 17: 0.6215 - 18: 0.2339 - Max F1: 0.3903 - Mean F1: 0.3971 ## 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 | |:-------------:|:-----:|:-----:|:---------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:------------:|:-------------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:-----------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:------:|:-------:| | 1.1311 | 1.0 | 5595 | 0.8741 | 0.0694 | 0.0845 | 0.1012 | 0.1184 | 0.1361 | 0.1543 | 0.1721 | 0.1896 | 0.2072 | 0.2254 | 0.2452 | 0.2638 | 0.2836 | 0.3001 | 0.3161 | 0.3204 | 0.3019 | 0.2198 | 0.75 | 0.8 | 0.85 | 0.9 | 0.7 | 0.8 | 0.85 | 0.85 | 0.8 | 0.8 | 0.95 | 0.8 | 0.85 | 0.9 | 0.9 | 0.9 | 0.85 | 0.95 | 0.9 | 0.0977 | 0.2012 | 0.3069 | 0.2180 | 0.3982 | 0.4146 | 0.4235 | 0.3110 | 0.3433 | 0.5029 | 0.5039 | 0.5275 | 0.2241 | 0.1802 | 0.3434 | 0.2343 | 0.3988 | 0.6105 | 0.2014 | 0.3204 | 0.3390 | | 0.7682 | 2.0 | 11190 | 0.8513 | 0.0938 | 0.1227 | 0.1492 | 0.1724 | 0.1944 | 0.2135 | 0.2336 | 0.2515 | 0.2706 | 0.2880 | 0.3058 | 0.3210 | 0.3374 | 0.3576 | 0.3733 | 0.3780 | 0.3697 | 0.3019 | 0.8 | 0.85 | 0.85 | 0.9 | 0.8 | 0.9 | 0.8 | 0.9 | 0.9 | 0.8 | 0.9 | 0.85 | 0.9 | 0.8 | 0.85 | 0.9 | 0.85 | 0.9 | 0.9 | 0.1535 | 0.3002 | 0.3611 | 0.3365 | 0.4672 | 0.4768 | 0.4414 | 0.3609 | 0.3684 | 0.5407 | 0.5423 | 0.5455 | 0.2423 | 0.1915 | 0.3768 | 0.3296 | 0.4296 | 0.6282 | 0.2284 | 0.3780 | 0.3853 | | 0.606 | 3.0 | 16785 | 0.8569 | 0.1078 | 0.1410 | 0.1700 | 0.1957 | 0.2179 | 0.2383 | 0.2580 | 0.2757 | 0.2935 | 0.3110 | 0.3275 | 0.3425 | 0.3592 | 0.3717 | 0.3829 | 0.3903 | 0.3847 | 0.3225 | 0.85 | 0.8 | 0.9 | 0.9 | 0.8 | 0.8 | 0.8 | 0.9 | 0.85 | 0.8 | 0.9 | 0.85 | 0.9 | 0.85 | 0.85 | 0.9 | 0.85 | 0.9 | 0.9 | 0.1654 | 0.3112 | 0.3764 | 0.3436 | 0.4800 | 0.4880 | 0.4593 | 0.3694 | 0.3882 | 0.5533 | 0.5439 | 0.5492 | 0.2443 | 0.2278 | 0.4014 | 0.3373 | 0.4511 | 0.6215 | 0.2339 | 0.3903 | 0.3971 | ### Framework versions - Transformers 4.39.3 - Pytorch 2.2.2+cu121 - Datasets 2.18.0 - Tokenizers 0.15.2