--- license: mit tags: - generated_from_trainer metrics: - precision - recall - f1 - accuracy model-index: - name: pos_final_xlm_nl results: [] --- # pos_final_xlm_nl This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.1066 - Precision: 0.9780 - Recall: 0.9783 - F1: 0.9782 - Accuracy: 0.9789 ## 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: 256 - eval_batch_size: 256 - seed: 42 - gradient_accumulation_steps: 4 - total_train_batch_size: 1024 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 500 - num_epochs: 40.0 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| | No log | 1.0 | 69 | 3.4837 | 0.2936 | 0.1709 | 0.2161 | 0.3200 | | No log | 2.0 | 138 | 0.8299 | 0.8501 | 0.8416 | 0.8459 | 0.8497 | | No log | 3.0 | 207 | 0.2765 | 0.9419 | 0.9408 | 0.9414 | 0.9429 | | No log | 4.0 | 276 | 0.1704 | 0.9601 | 0.9596 | 0.9599 | 0.9611 | | No log | 5.0 | 345 | 0.1259 | 0.9685 | 0.9686 | 0.9686 | 0.9693 | | No log | 6.0 | 414 | 0.1085 | 0.9711 | 0.9713 | 0.9712 | 0.9719 | | No log | 7.0 | 483 | 0.0984 | 0.9728 | 0.9731 | 0.9729 | 0.9738 | | 1.1448 | 8.0 | 552 | 0.0906 | 0.9742 | 0.9745 | 0.9743 | 0.9752 | | 1.1448 | 9.0 | 621 | 0.0888 | 0.9749 | 0.9752 | 0.9751 | 0.9758 | | 1.1448 | 10.0 | 690 | 0.0864 | 0.9757 | 0.9759 | 0.9758 | 0.9765 | | 1.1448 | 11.0 | 759 | 0.0842 | 0.9764 | 0.9767 | 0.9765 | 0.9772 | | 1.1448 | 12.0 | 828 | 0.0840 | 0.9764 | 0.9768 | 0.9766 | 0.9773 | | 1.1448 | 13.0 | 897 | 0.0846 | 0.9766 | 0.9769 | 0.9768 | 0.9775 | | 1.1448 | 14.0 | 966 | 0.0854 | 0.9768 | 0.9771 | 0.9769 | 0.9776 | | 0.0668 | 15.0 | 1035 | 0.0867 | 0.9767 | 0.9770 | 0.9768 | 0.9776 | | 0.0668 | 16.0 | 1104 | 0.0859 | 0.9769 | 0.9772 | 0.9771 | 0.9778 | | 0.0668 | 17.0 | 1173 | 0.0858 | 0.9772 | 0.9775 | 0.9773 | 0.9781 | | 0.0668 | 18.0 | 1242 | 0.0878 | 0.9776 | 0.9779 | 0.9778 | 0.9785 | | 0.0668 | 19.0 | 1311 | 0.0887 | 0.9775 | 0.9779 | 0.9777 | 0.9785 | | 0.0668 | 20.0 | 1380 | 0.0902 | 0.9774 | 0.9777 | 0.9775 | 0.9783 | | 0.0668 | 21.0 | 1449 | 0.0910 | 0.9772 | 0.9775 | 0.9774 | 0.9782 | | 0.0375 | 22.0 | 1518 | 0.0926 | 0.9774 | 0.9777 | 0.9775 | 0.9783 | | 0.0375 | 23.0 | 1587 | 0.0930 | 0.9777 | 0.9780 | 0.9779 | 0.9787 | | 0.0375 | 24.0 | 1656 | 0.0955 | 0.9777 | 0.9781 | 0.9779 | 0.9787 | | 0.0375 | 25.0 | 1725 | 0.0955 | 0.9778 | 0.9781 | 0.9780 | 0.9787 | | 0.0375 | 26.0 | 1794 | 0.0978 | 0.9776 | 0.9779 | 0.9777 | 0.9785 | | 0.0375 | 27.0 | 1863 | 0.0997 | 0.9772 | 0.9775 | 0.9774 | 0.9782 | | 0.0375 | 28.0 | 1932 | 0.1000 | 0.9776 | 0.9779 | 0.9778 | 0.9786 | | 0.0238 | 29.0 | 2001 | 0.1022 | 0.9775 | 0.9778 | 0.9776 | 0.9785 | | 0.0238 | 30.0 | 2070 | 0.1030 | 0.9777 | 0.9780 | 0.9779 | 0.9787 | | 0.0238 | 31.0 | 2139 | 0.1041 | 0.9778 | 0.9780 | 0.9779 | 0.9787 | | 0.0238 | 32.0 | 2208 | 0.1054 | 0.9778 | 0.9781 | 0.9779 | 0.9787 | | 0.0238 | 33.0 | 2277 | 0.1055 | 0.9777 | 0.9779 | 0.9778 | 0.9786 | | 0.0238 | 34.0 | 2346 | 0.1063 | 0.9778 | 0.9780 | 0.9779 | 0.9787 | | 0.0238 | 35.0 | 2415 | 0.1066 | 0.9780 | 0.9783 | 0.9782 | 0.9789 | | 0.0238 | 36.0 | 2484 | 0.1075 | 0.9779 | 0.9781 | 0.9780 | 0.9788 | | 0.0167 | 37.0 | 2553 | 0.1083 | 0.9780 | 0.9783 | 0.9781 | 0.9789 | | 0.0167 | 38.0 | 2622 | 0.1083 | 0.9780 | 0.9783 | 0.9781 | 0.9789 | | 0.0167 | 39.0 | 2691 | 0.1087 | 0.9779 | 0.9782 | 0.9781 | 0.9789 | | 0.0167 | 40.0 | 2760 | 0.1088 | 0.9780 | 0.9782 | 0.9781 | 0.9789 | ### Framework versions - Transformers 4.25.1 - Pytorch 1.12.0 - Datasets 2.18.0 - Tokenizers 0.13.2