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

Train Loss: 0.9863
Validation Loss: 1.1352
BLUE: 37.083

Intended uses & limitations

This model has been created for learning purposes at the MIARFID Automatic Translation course.

Training and evaluation data

This model was fine-tuned with a subset of the Europarl-v7-es-en, consisting of 50.000 sentences in English and Spanish.

Philipp Koehn. 2005. Europarl: A Parallel Corpus for Statistical Machine Translation. In Proceedings of Machine Translation Summit X: Papers, pages 79–86, Phuket, Thailand.

Training hyperparameters

The following hyperparameters were used during training:

optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-05, 'decay_steps': 4344, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
training_precision: mixed_float16

Training results

Train Loss	Validation Loss	Epoch
1.2441	1.1487	0
1.0785	1.1351	1
0.9863	1.1352	2

Framework versions

Transformers 4.37.0
TensorFlow 2.13.0
Datasets 2.16.1
Tokenizers 0.15.1