BARTBana_Translation_v00

This model is a fine-tuned version of vinai/bartpho-syllable on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6020
• Sacrebleu: 7.9509

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: 64
• eval_batch_size: 64
• 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
• num_epochs: 15
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Sacrebleu
1.2433	1.0	742	0.8630	0.8963
1.0005	2.0	1484	0.8077	1.7612
0.9144	3.0	2226	0.7620	2.8192
0.8945	4.0	2968	0.7246	3.6968
0.8435	5.0	3710	0.6943	4.8346
0.8251	6.0	4452	0.6687	5.5850
0.7872	7.0	5194	0.6513	6.1145
0.7674	8.0	5936	0.6385	6.5998
0.76	9.0	6678	0.6263	7.0741
0.7407	10.0	7420	0.6198	7.3214
0.7301	11.0	8162	0.6100	7.5060
0.7294	12.0	8904	0.6077	7.6978
0.7174	13.0	9646	0.6032	7.8715
0.7131	14.0	10388	0.6020	7.9509
0.7028	15.0	11130	0.6015	7.9382

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0