mabart-translation_en_es

This model is a fine-tuned version of facebook/mbart-large-50 on the dataset "Thermostatic/texts_parallel_corpus_europarl_english_spanish". It achieves the following results on the evaluation set:

• Loss: 1.3080
• Bleu: 27.9907
• Gen Len: 90.0233

Model description

This small model was developed to translate the minutes of the European Parliament proceedings from English into Spanish as part of a Master’s degree project in Natural Language Processing (NLP).

Intended uses & limitations

The model was trained on a relatively small dataset (3,000 rows), so its performance is limited. It is intended primarily for educational and experimental purposes.

Training and evaluation data

The model was trained on this dataset, which includes a large parallel corpus of English–Spanish sentence pairs: https://huggingface.co/datasets/Thermostatic/texts_parallel_corpus_europarl_english_spanish

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5.6e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Use OptimizerNames.ADAFACTOR and the args are: No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
1.4951	1.0	2700	1.3531	26.531	90.0033
0.8465	2.0	5400	1.3080	27.9907	90.0233

Framework versions

• Transformers 5.2.0
• Pytorch 2.10.0+cu128
• Datasets 4.6.1
• Tokenizers 0.22.2