--- library_name: transformers license: apache-2.0 base_model: Helsinki-NLP/opus-mt-es-es tags: - generated_from_trainer model-index: - name: 5_6kmslsamples results: [] --- # 5_6kmslsamples This model is a fine-tuned version of [Helsinki-NLP/opus-mt-es-es](https://huggingface.co/Helsinki-NLP/opus-mt-es-es) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.2686 - Bleu Msl: 79.6101 - Bleu Asl: 87.4768 - Ter Msl: 13.4848 - Ter Asl: 6.4259 ## 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: 0.0001 - train_batch_size: 32 - eval_batch_size: 64 - 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: 30 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Bleu Msl | Bleu Asl | Ter Msl | Ter Asl | |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|:-------:| | No log | 1.0 | 142 | 0.4517 | 4.3556 | 60.0932 | 263.6364 | 17.9336 | | No log | 2.0 | 284 | 0.2911 | 23.2666 | 85.6532 | 55.0000 | 7.1819 | | No log | 3.0 | 426 | 0.2635 | 7.6866 | 80.4693 | 145.5556 | 8.7358 | | 0.5021 | 4.0 | 568 | 0.2526 | 36.1099 | 86.3805 | 21.6162 | 7.3079 | | 0.5021 | 5.0 | 710 | 0.2554 | 29.9719 | 84.4134 | 24.9495 | 8.1898 | | 0.5021 | 6.0 | 852 | 0.2573 | 73.4067 | 87.3570 | 15.9091 | 6.5939 | | 0.5021 | 7.0 | 994 | 0.2607 | 75.9042 | 87.5621 | 14.0404 | 5.9219 | | 0.0602 | 8.0 | 1136 | 0.2534 | 69.4208 | 82.2155 | 17.9798 | 7.8118 | | 0.0602 | 9.0 | 1278 | 0.2540 | 78.9796 | 84.5857 | 13.6869 | 7.2659 | | 0.0602 | 10.0 | 1420 | 0.2439 | 78.3772 | 85.0526 | 14.0909 | 7.6438 | | 0.0277 | 11.0 | 1562 | 0.2586 | 79.0479 | 85.5579 | 13.5354 | 6.8879 | | 0.0277 | 12.0 | 1704 | 0.2658 | 77.8620 | 84.0916 | 14.3939 | 9.1978 | | 0.0277 | 13.0 | 1846 | 0.2593 | 78.6031 | 87.9318 | 12.8283 | 5.9219 | | 0.0277 | 14.0 | 1988 | 0.2574 | 77.2537 | 86.9300 | 13.6364 | 6.6779 | | 0.0173 | 15.0 | 2130 | 0.2599 | 77.3620 | 85.5624 | 14.6465 | 7.2239 | | 0.0173 | 16.0 | 2272 | 0.2674 | 77.8456 | 85.5155 | 15.0 | 7.3499 | | 0.0173 | 17.0 | 2414 | 0.2603 | 77.5280 | 86.8876 | 14.5960 | 7.1399 | | 0.0104 | 18.0 | 2556 | 0.2718 | 77.0685 | 86.1825 | 15.1515 | 6.5939 | | 0.0104 | 19.0 | 2698 | 0.2675 | 78.5651 | 86.7494 | 13.8384 | 6.5099 | | 0.0104 | 20.0 | 2840 | 0.2634 | 79.1140 | 85.3241 | 13.5859 | 7.3919 | | 0.0104 | 21.0 | 2982 | 0.2617 | 78.1030 | 86.0924 | 14.8990 | 7.0139 | | 0.0076 | 22.0 | 3124 | 0.2625 | 78.8043 | 86.6405 | 13.7374 | 6.2999 | | 0.0076 | 23.0 | 3266 | 0.2694 | 78.2577 | 86.4381 | 14.4444 | 6.2579 | | 0.0076 | 24.0 | 3408 | 0.2671 | 78.3892 | 86.1058 | 14.1919 | 6.5099 | | 0.0056 | 25.0 | 3550 | 0.2666 | 78.4721 | 86.1346 | 14.8990 | 6.7619 | | 0.0056 | 26.0 | 3692 | 0.2670 | 79.2248 | 87.4447 | 13.9899 | 6.4259 | | 0.0056 | 27.0 | 3834 | 0.2689 | 80.1481 | 87.0403 | 13.4343 | 6.4259 | | 0.0056 | 28.0 | 3976 | 0.2688 | 80.0229 | 87.3402 | 13.5354 | 6.2159 | | 0.005 | 29.0 | 4118 | 0.2688 | 79.5101 | 87.5201 | 13.5354 | 6.3839 | | 0.005 | 30.0 | 4260 | 0.2686 | 79.6101 | 87.4768 | 13.4848 | 6.4259 | ### Framework versions - Transformers 4.49.0 - Pytorch 2.5.1+cu124 - Datasets 3.3.2 - Tokenizers 0.21.0