exp2_10partition_modelo_asl3000

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: 0.1268
• Model Preparation Time: 0.0032
• 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: 0.9718
• Bleu 2 Asl: 0.9502
• Bleu 3 Asl: 0.9262
• Bleu 4 Asl: 0.8971
• Ter Asl: 3.1484

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	75	0.2877	0.0032	0	0	0	0	0	100	0	0.9471	0.9143	0.8823	0.8467	6.5089
No log	2.0	150	0.1600	0.0032	0	0	0	0	0	100	0	0.9560	0.9262	0.8951	0.8600	5.6213
No log	3.0	225	0.1465	0.0032	0	0	0	0	0	100	0	0.9625	0.9350	0.9076	0.8760	4.7337
No log	4.0	300	0.1421	0.0032	0	0	0	0	0	100	0	0.9670	0.9445	0.9201	0.8903	4.0680
No log	5.0	375	0.1441	0.0032	0	0	0	0	0	100	0	0.9607	0.9355	0.9092	0.8786	4.8077
No log	6.0	450	0.1509	0.0032	0	0	0	0	0	100	0	0.9612	0.9368	0.9113	0.8811	4.8817
0.2184	7.0	525	0.1463	0.0032	0	0	0	0	0	100	0	0.9651	0.9412	0.9155	0.8841	4.2160
0.2184	8.0	600	0.1574	0.0032	0	0	0	0	0	100	0	0.9430	0.9164	0.8878	0.8535	7.0266
0.2184	9.0	675	0.1478	0.0032	0	0	0	0	0	100	0	0.9628	0.9401	0.9157	0.8855	4.4379
0.2184	10.0	750	0.1471	0.0032	0	0	0	0	0	100	0	0.9659	0.9437	0.9198	0.8901	4.0680
0.2184	11.0	825	0.1511	0.0032	0	0	0	0	0	100	0	0.9658	0.9428	0.9183	0.8881	4.2899
0.2184	12.0	900	0.1446	0.0032	0	0	0	0	0	100	0	0.9665	0.9428	0.9177	0.8867	3.9201
0.2184	13.0	975	0.1470	0.0032	0	0	0	0	0	100	0	0.9683	0.9456	0.9218	0.8934	3.8462
0.013	14.0	1050	0.1504	0.0032	0	0	0	0	0	100	0	0.9671	0.9431	0.9179	0.8876	3.8462
0.013	15.0	1125	0.1497	0.0032	0	0	0	0	0	100	0	0.9678	0.9438	0.9180	0.8873	3.7722
0.013	16.0	1200	0.1616	0.0032	0	0	0	0	0	100	0	0.9652	0.9409	0.9152	0.8845	4.1420
0.013	17.0	1275	0.1578	0.0032	0	0	0	0	0	100	0	0.9671	0.9447	0.9204	0.8913	3.8462
0.013	18.0	1350	0.1534	0.0032	0	0	0	0	0	100	0	0.9659	0.9425	0.9183	0.8890	3.9941
0.013	19.0	1425	0.1526	0.0032	0	0	0	0	0	100	0	0.9677	0.9453	0.9217	0.8939	3.6982
0.0058	20.0	1500	0.1574	0.0032	0	0	0	0	0	100	0	0.9684	0.9466	0.9230	0.8939	3.6243
0.0058	21.0	1575	0.1563	0.0032	0	0	0	0	0	100	0	0.9684	0.9469	0.9242	0.8961	3.7722
0.0058	22.0	1650	0.1575	0.0032	0	0	0	0	0	100	0	0.9690	0.9476	0.9249	0.8970	3.6243
0.0058	23.0	1725	0.1589	0.0032	0	0	0	0	0	100	0	0.9678	0.9459	0.9223	0.8931	3.6243
0.0058	24.0	1800	0.1628	0.0032	0	0	0	0	0	100	0	0.9671	0.9451	0.9214	0.8920	3.7722
0.0058	25.0	1875	0.1576	0.0032	0	0	0	0	0	100	0	0.9678	0.9454	0.9216	0.8922	3.6982
0.0058	26.0	1950	0.1564	0.0032	0	0	0	0	0	100	0	0.9697	0.9480	0.9246	0.8958	3.4024
0.0029	27.0	2025	0.1555	0.0032	0	0	0	0	0	100	0	0.9697	0.9480	0.9246	0.8957	3.4763
0.0029	28.0	2100	0.1561	0.0032	0	0	0	0	0	100	0	0.9691	0.9469	0.9229	0.8935	3.5503
0.0029	29.0	2175	0.1559	0.0032	0	0	0	0	0	100	0	0.9697	0.9480	0.9246	0.8955	3.4763
0.0029	30.0	2250	0.1561	0.0032	0	0	0	0	0	100	0	0.9691	0.9469	0.9229	0.8935	3.5503

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1