exp3_10partition_modelo_asl6000

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: 0.1360
• Model Preparation Time: 0.0034
• Bleu Msl: 0
• Bleu 1 Msl: 0
• Bleu 2 Msl: 0
• Bleu 3 Msl: 0
• Bleu 4 Msl: 0
• Ter Msl: 100
• Bleu Asl: 0
• Bleu 1 Asl: 0.9771
• Bleu 2 Asl: 0.9588
• Bleu 3 Asl: 0.9406
• Bleu 4 Asl: 0.9197
• Ter Asl: 2.9116

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	Model Preparation Time	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	150	0.1715	0.0034	0	0	0	0	0	100	0	0.9510	0.9205	0.8886	0.8535	5.9367
No log	2.0	300	0.1202	0.0034	0	0	0	0	0	100	0	0.9683	0.9475	0.9243	0.8963	4.0882
No log	3.0	450	0.1128	0.0034	0	0	0	0	0	100	0	0.9689	0.9476	0.9237	0.8956	3.9104
0.2629	4.0	600	0.1179	0.0034	0	0	0	0	0	100	0	0.9714	0.9519	0.9288	0.9022	3.4838
0.2629	5.0	750	0.1103	0.0034	0	0	0	0	0	100	0	0.9715	0.9523	0.9297	0.9022	3.5194
0.2629	6.0	900	0.1103	0.0034	0	0	0	0	0	100	0	0.9724	0.9525	0.9288	0.9005	3.4483
0.0327	7.0	1050	0.1116	0.0034	0	0	0	0	0	100	0	0.9705	0.9499	0.9273	0.9002	3.7327
0.0327	8.0	1200	0.1067	0.0034	0	0	0	0	0	100	0	0.9736	0.9542	0.9311	0.9039	3.2705
0.0327	9.0	1350	0.1060	0.0034	0	0	0	0	0	100	0	0.9739	0.9554	0.9334	0.9068	3.2705
0.0142	10.0	1500	0.1117	0.0034	0	0	0	0	0	100	0	0.9743	0.9559	0.9354	0.9103	3.1639
0.0142	11.0	1650	0.1044	0.0034	0	0	0	0	0	100	0	0.9740	0.9556	0.9341	0.9091	3.0217
0.0142	12.0	1800	0.1055	0.0034	0	0	0	0	0	100	0	0.9768	0.9595	0.9383	0.9124	2.8084
0.0142	13.0	1950	0.1005	0.0034	0	0	0	0	0	100	0	0.9793	0.9636	0.9446	0.9213	2.5240
0.0094	14.0	2100	0.1058	0.0034	0	0	0	0	0	100	0	0.9752	0.9574	0.9356	0.9098	2.9150
0.0094	15.0	2250	0.1161	0.0034	0	0	0	0	0	100	0	0.9765	0.9589	0.9372	0.9110	2.8795
0.0094	16.0	2400	0.1240	0.0034	0	0	0	0	0	100	0	0.9730	0.9538	0.9308	0.9037	3.3416
0.0067	17.0	2550	0.1137	0.0034	0	0	0	0	0	100	0	0.9746	0.9563	0.9339	0.9074	3.0572
0.0067	18.0	2700	0.1157	0.0034	0	0	0	0	0	100	0	0.9752	0.9570	0.9348	0.9085	2.9150
0.0067	19.0	2850	0.1068	0.0034	0	0	0	0	0	100	0	0.9768	0.9589	0.9373	0.9117	2.7728
0.0042	20.0	3000	0.1111	0.0034	0	0	0	0	0	100	0	0.9771	0.9598	0.9387	0.9133	2.7017
0.0042	21.0	3150	0.1073	0.0034	0	0	0	0	0	100	0	0.9746	0.9565	0.9350	0.9093	2.9506
0.0042	22.0	3300	0.1097	0.0034	0	0	0	0	0	100	0	0.9756	0.9582	0.9372	0.9118	2.8439
0.0042	23.0	3450	0.1132	0.0034	0	0	0	0	0	100	0	0.9780	0.9612	0.9407	0.9161	2.6662
0.0029	24.0	3600	0.1119	0.0034	0	0	0	0	0	100	0	0.9777	0.9607	0.9399	0.9148	2.7373
0.0029	25.0	3750	0.1105	0.0034	0	0	0	0	0	100	0	0.9771	0.9607	0.9404	0.9161	2.7373
0.0029	26.0	3900	0.1104	0.0034	0	0	0	0	0	100	0	0.9774	0.9608	0.9403	0.9154	2.7373
0.0017	27.0	4050	0.1103	0.0034	0	0	0	0	0	100	0	0.9777	0.9613	0.9409	0.9162	2.6662
0.0017	28.0	4200	0.1102	0.0034	0	0	0	0	0	100	0	0.9774	0.9606	0.9396	0.9144	2.7017
0.0017	29.0	4350	0.1109	0.0034	0	0	0	0	0	100	0	0.9771	0.9602	0.9392	0.9139	2.7373
0.0013	30.0	4500	0.1107	0.0034	0	0	0	0	0	100	0	0.9768	0.9597	0.9384	0.9126	2.7728

Framework versions

• Transformers 4.50.2
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1