En-Tn_update

This model is a fine-tuned version of Helsinki-NLP/opus-mt-en-tn on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.13002
• Bleu: 39.1470

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: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Epoch	Training Loss	Validation Loss	Bleu
1	1.929300	1.884056	29.762382
2	1.637300	1.605588	32.846868
3	1.500000	1.457442	34.307484
4	1.402400	1.356578	35.423774
5	1.324000	1.276492	36.553368
6	1.251300	1.221768	37.464270
7	1.224700	1.181320	38.157490
8	1.193200	1.152997	38.800566
9	1.166700	1.136147	38.985707
10	1.142500	1.130020	39.209327

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1