vinai_v2_meddict_meddictgem_2

This model is a fine-tuned version of vinai/vinai-translate-en2vi-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6615
• Bleu: 61.6514
• Gen Len: 43.4462

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: 128
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 5.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
0.7417	1.0	1004	0.6186	59.81	43.4055
0.4372	2.0	2008	0.6011	61.1202	43.4623
0.3542	3.0	3012	0.6146	61.528	43.4571
0.3375	4.0	4016	0.6407	61.5836	43.4415
0.3821	5.0	5020	0.6615	61.6514	43.4462

Framework versions

• Transformers 4.54.0
• Pytorch 2.7.0
• Datasets 4.1.1
• Tokenizers 0.21.4