opus-mt-ar-en-finetuned_augmented_synthetic-ar-to-en

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

• Loss: 0.8682
• Bleu: 63.4498
• Gen Len: 59.457

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.9549	1.0	1105	1.2644	43.0637	61.33
0.7674	2.0	2210	1.0862	51.6055	60.714
0.6736	3.0	3315	0.9910	56.1642	60.434
0.6011	4.0	4420	0.9463	59.6059	59.682
0.5543	5.0	5525	0.9158	61.101	59.493
0.5176	6.0	6630	0.8961	61.9065	59.468
0.4849	7.0	7735	0.8840	62.6833	59.5
0.4692	8.0	8840	0.8727	63.0766	59.425
0.464	9.0	9945	0.8709	63.3354	59.454
0.4486	10.0	11050	0.8682	63.4498	59.457

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3