Ts-En_update

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

• Loss: 1.2030
• Bleu: 44.6835

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	1.650900	1.676816	31.785027
2	1.365800	1.478832	35.611479
3	1.205400	1.376320	38.719939
4	1.095300	1.315572	40.419038
5	1.020000	1.275409	41.858332
6	0.959000	1.245399	42.456983
7	0.915700	1.225268	41.888945
8	0.879100	1.213904	43.738043
9	0.841900	1.205976	43.683633
10	0.800600	1.203017	44.294833

Framework versions

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