Vietnamese_Accent_AI_10000

This model is a fine-tuned version of NlpHUST/gpt2-vietnamese on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 4.3709
• Bleu: 40.2232

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu
3.4573	0.5	1000	4.3933	40.0227
4.1146	1.0	2000	4.2023	39.8481
3.485	1.5	3000	4.2376	39.9626
3.5035	2.0	4000	4.2281	39.9341
3.2949	2.5	5000	4.2746	39.9964
3.2822	3.0	6000	4.2468	40.1508
3.0944	3.5	7000	4.3169	40.2196
3.0728	4.0	8000	4.3168	40.2273
2.9938	4.5	9000	4.3716	40.2458
2.9584	5.0	10000	4.3709	40.2232

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1