fine-tune-phobert-phone-sentiment

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

• Loss: 0.5014
• Accuracy: 0.745
• Auc: 0.835

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc
0.6699	1.0	298	0.6424	0.666	0.771
0.6407	2.0	596	0.6114	0.701	0.797
0.6117	3.0	894	0.5846	0.72	0.804
0.5994	4.0	1192	0.5650	0.733	0.814
0.5836	5.0	1490	0.5511	0.736	0.818
0.5665	6.0	1788	0.5444	0.726	0.821
0.5645	7.0	2086	0.5317	0.736	0.824
0.5536	8.0	2384	0.5270	0.74	0.824
0.556	9.0	2682	0.5213	0.74	0.825
0.5463	10.0	2980	0.5174	0.737	0.829
0.5502	11.0	3278	0.5126	0.747	0.831
0.5434	12.0	3576	0.5096	0.743	0.831
0.5426	13.0	3874	0.5087	0.743	0.831
0.5347	14.0	4172	0.5095	0.739	0.831
0.5468	15.0	4470	0.5059	0.743	0.832
0.5364	16.0	4768	0.5038	0.743	0.833
0.5422	17.0	5066	0.5039	0.745	0.834
0.539	18.0	5364	0.5021	0.749	0.835
0.5237	19.0	5662	0.5016	0.743	0.835
0.5232	20.0	5960	0.5014	0.745	0.835

Framework versions

• Transformers 4.49.0
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0