mbart-large-50-en-es-translation-lr-1e-05-weight-decay-0.1

This model is a fine-tuned version of facebook/mbart-large-50 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9532
• Bleu: 45.1551
• Rouge: {'rouge1': 0.707093830119779, 'rouge2': 0.5240989044660875, 'rougeL': 0.6865395711179825, 'rougeLsum': 0.6867643949864491}

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Rouge
1.4485	1.0	4500	1.0236	42.1586	{'rouge1': 0.6728104679322686, 'rouge2': 0.4866267759088613, 'rougeL': 0.6507619922873461, 'rougeLsum': 0.6508024989844624}
0.8867	2.0	9000	0.9542	44.1945	{'rouge1': 0.6933374960151913, 'rouge2': 0.5090654274262618, 'rougeL': 0.6722360570050694, 'rougeLsum': 0.6723972406375381}
0.7112	3.0	13500	0.9408	44.9173	{'rouge1': 0.7047659807760827, 'rouge2': 0.5200169348076622, 'rougeL': 0.6839031690668775, 'rougeLsum': 0.6842067045539153}
0.6075	4.0	18000	0.9532	45.2020	{'rouge1': 0.7070170730434684, 'rouge2': 0.5239391023023636, 'rougeL': 0.6863309446860562, 'rougeLsum': 0.6866635686411662}

Framework versions

• Transformers 4.33.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4.dev0
• Tokenizers 0.13.3