Revision_PhoBert_Lexical_MetaPho_Q1

This model is a fine-tuned version of phunganhsang/PhoBert_Lexical_Dataset45K on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3628
• Accuracy: 0.8712
• F1: 0.6504

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_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.3497	100	0.5219	0.7676	0.5370
No log	0.6993	200	0.3648	0.8452	0.6048
0.3317	1.0490	300	0.3408	0.8624	0.6311
0.3317	1.3986	400	0.3949	0.8364	0.6027
0.3317	1.7483	500	0.4566	0.8053	0.5879
0.2151	2.0979	600	0.3826	0.8518	0.6231
0.2151	2.4476	700	0.3385	0.8742	0.6425
0.2151	2.7972	800	0.3997	0.8487	0.6276
0.1610	3.1469	900	0.3860	0.8556	0.6354
0.1610	3.4965	1000	0.3598	0.8691	0.6496
0.1610	3.8462	1100	0.3614	0.8695	0.6488
0.1222	4.1958	1200	0.3683	0.8676	0.6477
0.1222	4.5455	1300	0.3678	0.8695	0.6497
0.1222	4.8951	1400	0.3628	0.8712	0.6504

Framework versions

• Transformers 5.3.0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2