opus-mt-en-ar-evaluated-en-to-ar-1000instancesopus-leaningRate2e-05-batchSize8-11epoch-3

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-ar on the opus100 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1421
• Bleu: 21.3028
• Meteor: 0.1285
• Gen Len: 9.975

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 11
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Meteor	Gen Len
1.0508	1.0	100	0.1413	27.9009	0.1416	8.85
0.1253	2.0	200	0.1372	23.11	0.1345	9.855
0.1017	3.0	300	0.1390	21.7885	0.1364	9.97
0.0868	4.0	400	0.1378	21.3889	0.1314	9.835
0.0754	5.0	500	0.1398	22.198	0.132	9.675
0.0667	6.0	600	0.1396	20.8645	0.1308	10.055
0.0604	7.0	700	0.1408	20.289	0.1303	10.53
0.0553	8.0	800	0.1414	21.7023	0.1293	10.005
0.0518	9.0	900	0.1421	21.3028	0.1285	9.975

Framework versions

• Transformers 4.18.0
• Pytorch 1.11.0
• Datasets 2.1.0
• Tokenizers 0.12.1