PhoBert_Hosting_Dataset65K

This model is a fine-tuned version of vinai/phobert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4987
• Accuracy: 0.8753
• F1: 0.8746

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

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	0.2326	200	0.3396	0.8482	0.8465
No log	0.4651	400	0.3066	0.8666	0.8659
No log	0.6977	600	0.3778	0.8363	0.8361
No log	0.9302	800	0.3028	0.8668	0.8655
0.3824	1.1628	1000	0.2929	0.8765	0.8756
0.3824	1.3953	1200	0.3120	0.8690	0.8687
0.3824	1.6279	1400	0.2890	0.8778	0.8772
0.3824	1.8605	1600	0.2843	0.8780	0.8775
0.3043	2.0930	1800	0.2801	0.8803	0.8797
0.3043	2.3256	2000	0.3004	0.8759	0.8756
0.3043	2.5581	2200	0.2884	0.8772	0.8763
0.3043	2.7907	2400	0.2813	0.8813	0.8807
0.269	3.0233	2600	0.2837	0.8794	0.8790
0.269	3.2558	2800	0.2957	0.8802	0.8796
0.269	3.4884	3000	0.3030	0.8801	0.8791
0.269	3.7209	3200	0.2750	0.8804	0.8799
0.269	3.9535	3400	0.2860	0.8795	0.8793
0.2393	4.1860	3600	0.3115	0.8819	0.8810
0.2393	4.4186	3800	0.2944	0.8836	0.8832
0.2393	4.6512	4000	0.3004	0.8752	0.8749
0.2393	4.8837	4200	0.2992	0.8800	0.8791
0.2121	5.1163	4400	0.3067	0.8787	0.8780
0.2121	5.3488	4600	0.3176	0.8811	0.8803
0.2121	5.5814	4800	0.3175	0.8825	0.8822
0.2121	5.8140	5000	0.2999	0.8780	0.8776
0.185	6.0465	5200	0.3545	0.8731	0.8728
0.185	6.2791	5400	0.3177	0.8809	0.8804
0.185	6.5116	5600	0.3344	0.8781	0.8777
0.185	6.7442	5800	0.3397	0.8762	0.8759
0.185	6.9767	6000	0.3447	0.8807	0.8799
0.1603	7.2093	6200	0.3804	0.8759	0.8756
0.1603	7.4419	6400	0.3587	0.8782	0.8776
0.1603	7.6744	6600	0.3743	0.8752	0.8749
0.1603	7.9070	6800	0.3755	0.8792	0.8788
0.1414	8.1395	7000	0.3865	0.8816	0.8806
0.1414	8.3721	7200	0.3805	0.8748	0.8738
0.1414	8.6047	7400	0.4066	0.8725	0.8722
0.1414	8.8372	7600	0.3818	0.8796	0.8791
0.1235	9.0698	7800	0.4175	0.8792	0.8786
0.1235	9.3023	8000	0.4086	0.8783	0.8776
0.1235	9.5349	8200	0.3963	0.8798	0.8790
0.1235	9.7674	8400	0.3993	0.8764	0.8757
0.1093	10.0	8600	0.3975	0.8748	0.8743
0.1093	10.2326	8800	0.4290	0.8755	0.8748
0.1093	10.4651	9000	0.4457	0.8756	0.8753
0.1093	10.6977	9200	0.4222	0.8749	0.8743
0.1093	10.9302	9400	0.4373	0.8733	0.8723
0.0967	11.1628	9600	0.4547	0.8779	0.8774
0.0967	11.3953	9800	0.4524	0.8766	0.8758
0.0967	11.6279	10000	0.4393	0.8751	0.8741
0.0967	11.8605	10200	0.4380	0.8768	0.8762
0.0883	12.0930	10400	0.4605	0.8771	0.8761
0.0883	12.3256	10600	0.4615	0.8756	0.8748
0.0883	12.5581	10800	0.4642	0.8747	0.8739
0.0883	12.7907	11000	0.4741	0.8743	0.8737
0.0796	13.0233	11200	0.4863	0.8761	0.8752
0.0796	13.2558	11400	0.4771	0.8759	0.8750
0.0796	13.4884	11600	0.4863	0.8772	0.8766
0.0796	13.7209	11800	0.4881	0.8744	0.8736
0.0796	13.9535	12000	0.4828	0.8748	0.8740
0.0746	14.1860	12200	0.4992	0.8761	0.8755
0.0746	14.4186	12400	0.4976	0.8763	0.8756
0.0746	14.6512	12600	0.4988	0.8752	0.8745
0.0746	14.8837	12800	0.4987	0.8753	0.8746

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1