BARTBana_v5

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: 1.5331

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: 8
• eval_batch_size: 8
• 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
3.1425	1.0	6519	2.7167
2.6259	2.0	13038	2.2817
2.3452	3.0	19557	2.1062
2.1811	4.0	26076	1.9017
2.0841	5.0	32595	1.8220
1.9879	6.0	39114	1.7374
1.9034	7.0	45633	1.7504
1.8565	8.0	52152	1.6704
1.8123	9.0	58671	1.6438
1.7974	10.0	65190	1.5835
1.7586	11.0	71709	1.5772
1.7371	12.0	78228	1.5353
1.7152	13.0	84747	1.5148
1.6921	14.0	91266	1.4985
1.688	15.0	97785	1.5129

Framework versions

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