modelo_ta_es_en_ddhh

This model is a fine-tuned version of Helsinki-NLP/opus-mt-es-en on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5067
• Bleu: 62.2593

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
0.8053	1.0	11	0.4976	59.4263
0.5226	2.0	22	0.4817	56.5988
0.4773	3.0	33	0.4812	58.7628
0.2781	4.0	44	0.4856	60.2774
0.3026	5.0	55	0.4914	60.1256
0.2559	6.0	66	0.4969	60.1256
0.2074	7.0	77	0.5001	60.3113
0.1935	8.0	88	0.5049	60.3113
0.1811	9.0	99	0.5061	62.2593
0.1420	10.0	110	0.5067	62.2593

Framework versions

• Transformers 5.0.0
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.22.2