exp5_10partition_modeloorig

This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-es on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3873
• Bleu Msl: 100.0000
• Bleu 1 Msl: 0.4433
• Bleu 2 Msl: 0.0122
• Bleu 3 Msl: 0.0039
• Bleu 4 Msl: 0.0020
• Ter Msl: {'score': 33.771929824561404, 'num_edits': 308, 'ref_length': 912.0}
• Bleu Asl: 0
• Bleu 1 Asl: 0
• Bleu 2 Asl: 0
• Bleu 3 Asl: 0
• Bleu 4 Asl: 0
• Ter Asl: 100

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 1 Msl	Bleu 2 Msl	Bleu 3 Msl	Bleu 4 Msl	Ter Msl	Bleu Asl	Bleu 1 Asl	Bleu 2 Asl	Bleu 3 Asl	Bleu 4 Asl	Ter Asl
No log	1.0	75	2.2027	35.3553	0.1600	0.0073	0.0028	0.0016	{'score': 1123.3552631578948, 'num_edits': 10245, 'ref_length': 912.0}	0	0	0	0	0	100
No log	2.0	150	1.5230	0.0	0.25	0.0091	0.0032	0.0018	{'score': 71.38157894736842, 'num_edits': 651, 'ref_length': 912.0}	0	0	0	0	0	100
No log	3.0	225	1.5794	14.0585	0.2433	0.0090	0.0032	0.0017	{'score': 61.622807017543856, 'num_edits': 562, 'ref_length': 912.0}	0	0	0	0	0	100
No log	4.0	300	1.5053	0.0	0.2967	0.0100	0.0034	0.0018	{'score': 59.97807017543859, 'num_edits': 547, 'ref_length': 912.0}	0	0	0	0	0	100
No log	5.0	375	1.4611	0.0	0.3067	0.0101	0.0034	0.0018	{'score': 47.368421052631575, 'num_edits': 432, 'ref_length': 912.0}	0	0	0	0	0	100
No log	6.0	450	1.4808	100.0000	0.2967	0.0100	0.0034	0.0018	{'score': 55.26315789473685, 'num_edits': 504, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	7.0	525	1.4529	100.0000	0.3933	0.0115	0.0037	0.0020	{'score': 42.65350877192983, 'num_edits': 389, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	8.0	600	1.3500	100.0000	0.3667	0.0111	0.0037	0.0019	{'score': 42.43421052631579, 'num_edits': 387, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	9.0	675	1.3970	0.0	0.3433	0.0107	0.0036	0.0019	{'score': 44.84649122807017, 'num_edits': 409, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	10.0	750	1.2626	100.0000	0.4467	0.0122	0.0039	0.0020	{'score': 35.526315789473685, 'num_edits': 324, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	11.0	825	1.3315	100.0000	0.41	0.0117	0.0038	0.0020	{'score': 36.18421052631579, 'num_edits': 330, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	12.0	900	1.3318	100.0000	0.38	0.0113	0.0037	0.0019	{'score': 41.00877192982456, 'num_edits': 374, 'ref_length': 912.0}	0	0	0	0	0	100
0.5768	13.0	975	1.3158	0.0	0.3867	0.0114	0.0037	0.0020	{'score': 42.98245614035088, 'num_edits': 392, 'ref_length': 912.0}	0	0	0	0	0	100
0.0447	14.0	1050	1.4628	100.0000	0.4133	0.0118	0.0038	0.0020	{'score': 39.03508771929825, 'num_edits': 356, 'ref_length': 912.0}	0	0	0	0	0	100
0.0447	15.0	1125	1.3666	100.0000	0.44	0.0121	0.0039	0.0020	{'score': 33.99122807017544, 'num_edits': 310, 'ref_length': 912.0}	0	0	0	0	0	100
0.0447	16.0	1200	1.3960	100.0000	0.4367	0.0121	0.0039	0.0020	{'score': 31.798245614035086, 'num_edits': 290, 'ref_length': 912.0}	0	0	0	0	0	100
0.0447	17.0	1275	1.3480	0.0	0.3867	0.0114	0.0037	0.0020	{'score': 38.81578947368421, 'num_edits': 354, 'ref_length': 912.0}	0	0	0	0	0	100
0.0447	18.0	1350	1.3485	100.0000	0.4533	0.0123	0.0039	0.0020	{'score': 33.44298245614035, 'num_edits': 305, 'ref_length': 912.0}	0	0	0	0	0	100
0.0447	19.0	1425	1.3811	100.0000	0.4233	0.0119	0.0038	0.0020	{'score': 35.74561403508772, 'num_edits': 326, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	20.0	1500	1.3515	100.0000	0.4367	0.0121	0.0039	0.0020	{'score': 34.75877192982456, 'num_edits': 317, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	21.0	1575	1.3758	100.0000	0.44	0.0121	0.0039	0.0020	{'score': 34.53947368421053, 'num_edits': 315, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	22.0	1650	1.3835	100.0000	0.4467	0.0122	0.0039	0.0020	{'score': 37.06140350877193, 'num_edits': 338, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	23.0	1725	1.3532	100.0000	0.4467	0.0122	0.0039	0.0020	{'score': 33.99122807017544, 'num_edits': 310, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	24.0	1800	1.4047	100.0000	0.4367	0.0121	0.0039	0.0020	{'score': 34.868421052631575, 'num_edits': 318, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	25.0	1875	1.3566	100.0000	0.4433	0.0122	0.0039	0.0020	{'score': 34.32017543859649, 'num_edits': 313, 'ref_length': 912.0}	0	0	0	0	0	100
0.0175	26.0	1950	1.4093	100.0000	0.4433	0.0122	0.0039	0.0020	{'score': 33.00438596491228, 'num_edits': 301, 'ref_length': 912.0}	0	0	0	0	0	100
0.0103	27.0	2025	1.3872	100.0000	0.4433	0.0122	0.0039	0.0020	{'score': 33.33333333333333, 'num_edits': 304, 'ref_length': 912.0}	0	0	0	0	0	100
0.0103	28.0	2100	1.3895	100.0000	0.4433	0.0122	0.0039	0.0020	{'score': 33.771929824561404, 'num_edits': 308, 'ref_length': 912.0}	0	0	0	0	0	100
0.0103	29.0	2175	1.3913	100.0000	0.4433	0.0122	0.0039	0.0020	{'score': 33.771929824561404, 'num_edits': 308, 'ref_length': 912.0}	0	0	0	0	0	100
0.0103	30.0	2250	1.3873	100.0000	0.4433	0.0122	0.0039	0.0020	{'score': 33.771929824561404, 'num_edits': 308, 'ref_length': 912.0}	0	0	0	0	0	100

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0