phoBert-vietnamese-nli

This model is a fine-tuned version of vinai/phobert-base-v2 on the lizNguyen235/vietnamese-nli-phobert dataset. It achieves the following results on the evaluation set:

• Loss: 0.6402
• Accuracy: 0.8259
• F1: 0.8257
• Precision: 0.8257
• Recall: 0.8259

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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.6467	0.0728	500	0.6709	0.7389	0.7388	0.7388	0.7389
0.6058	0.1455	1000	0.6508	0.7443	0.7458	0.7498	0.7443
0.5669	0.2183	1500	0.6069	0.7552	0.7550	0.7566	0.7552
0.5684	0.2911	2000	0.5917	0.7561	0.7543	0.7550	0.7561
0.5622	0.3638	2500	0.5752	0.7652	0.7651	0.7651	0.7652
0.536	0.4366	3000	0.5603	0.7833	0.7824	0.7823	0.7833
0.5304	0.5094	3500	0.5723	0.7779	0.7783	0.7812	0.7779
0.5109	0.5822	4000	0.5474	0.7842	0.7855	0.7883	0.7842
0.5272	0.6549	4500	0.5517	0.7860	0.7875	0.7935	0.7860
0.5194	0.7277	5000	0.5316	0.7978	0.7984	0.7995	0.7978
0.5129	0.8005	5500	0.5568	0.7797	0.7824	0.7927	0.7797
0.512	0.8732	6000	0.5316	0.7879	0.7891	0.7927	0.7879
0.5111	0.9460	6500	0.5116	0.8060	0.8062	0.8070	0.8060
0.4333	1.0188	7000	0.5270	0.8005	0.8002	0.8000	0.8005
0.4233	1.0915	7500	0.5719	0.7951	0.7939	0.7950	0.7951
0.4357	1.1643	8000	0.5319	0.8087	0.8089	0.8098	0.8087
0.4384	1.2371	8500	0.5206	0.8141	0.8147	0.8158	0.8141
0.4296	1.3099	9000	0.5182	0.7978	0.7976	0.7978	0.7978
0.4228	1.3826	9500	0.5163	0.8114	0.8121	0.8132	0.8114
0.442	1.4554	10000	0.5105	0.8051	0.8046	0.8051	0.8051
0.4325	1.5282	10500	0.5124	0.8114	0.8111	0.8113	0.8114
0.4295	1.6009	11000	0.4973	0.8214	0.8210	0.8224	0.8214
0.4163	1.6737	11500	0.4986	0.8169	0.8165	0.8164	0.8169
0.4073	1.7465	12000	0.5070	0.8141	0.8150	0.8168	0.8141
0.409	1.8192	12500	0.5330	0.7996	0.8002	0.8022	0.7996
0.4306	1.8920	13000	0.5204	0.8051	0.8059	0.8081	0.8051
0.4117	1.9648	13500	0.4972	0.8160	0.8145	0.8160	0.8160
0.3386	2.0375	14000	0.5470	0.8169	0.8180	0.8212	0.8169
0.3356	2.1103	14500	0.5304	0.8060	0.8053	0.8053	0.8060
0.3375	2.1831	15000	0.5376	0.8042	0.8036	0.8034	0.8042
0.3392	2.2559	15500	0.5408	0.8105	0.8105	0.8107	0.8105
0.3459	2.3286	16000	0.5239	0.8150	0.8154	0.8160	0.8150
0.3536	2.4014	16500	0.5321	0.8078	0.8092	0.8128	0.8078
0.3642	2.4742	17000	0.5210	0.8087	0.8082	0.8080	0.8087
0.3511	2.5469	17500	0.5229	0.8087	0.8070	0.8077	0.8087
0.3484	2.6197	18000	0.5318	0.8042	0.8046	0.8055	0.8042
0.3614	2.6925	18500	0.5260	0.8132	0.8142	0.8165	0.8132
0.3403	2.7652	19000	0.5296	0.8178	0.8174	0.8172	0.8178
0.3469	2.8380	19500	0.5125	0.8160	0.8165	0.8173	0.8160
0.3517	2.9108	20000	0.5108	0.8241	0.8242	0.8245	0.8241
0.3435	2.9836	20500	0.5074	0.8241	0.8242	0.8246	0.8241
0.2722	3.0563	21000	0.5576	0.8169	0.8178	0.8210	0.8169
0.2726	3.1291	21500	0.5900	0.8114	0.8131	0.8183	0.8114
0.2822	3.2019	22000	0.5513	0.8132	0.8142	0.8164	0.8132
0.2668	3.2746	22500	0.5661	0.8123	0.8118	0.8118	0.8123
0.2956	3.3474	23000	0.5287	0.8196	0.8201	0.8211	0.8196
0.2626	3.4202	23500	0.5841	0.8123	0.8123	0.8122	0.8123
0.2834	3.4929	24000	0.5570	0.8160	0.8158	0.8157	0.8160
0.2866	3.5657	24500	0.5544	0.8178	0.8180	0.8205	0.8178
0.2891	3.6385	25000	0.5380	0.8223	0.8232	0.8257	0.8223
0.2782	3.7113	25500	0.5622	0.8114	0.8124	0.8148	0.8114
0.2934	3.7840	26000	0.5426	0.8223	0.8221	0.8220	0.8223
0.2718	3.8568	26500	0.5858	0.8069	0.8077	0.8101	0.8069
0.2896	3.9296	27000	0.5363	0.8232	0.8233	0.8250	0.8232
0.2837	4.0023	27500	0.5829	0.8223	0.8219	0.8218	0.8223
0.2266	4.0751	28000	0.6141	0.8196	0.8193	0.8195	0.8196
0.2068	4.1479	28500	0.6231	0.8105	0.8101	0.8106	0.8105
0.2066	4.2206	29000	0.6119	0.8150	0.8153	0.8159	0.8150
0.2253	4.2934	29500	0.5815	0.8169	0.8174	0.8184	0.8169
0.2116	4.3662	30000	0.6242	0.8015	0.8025	0.8072	0.8015
0.2186	4.4389	30500	0.6570	0.8105	0.8112	0.8124	0.8105
0.2367	4.5117	31000	0.5854	0.8187	0.8185	0.8185	0.8187
0.2254	4.5845	31500	0.6053	0.8178	0.8184	0.8195	0.8178
0.2311	4.6573	32000	0.6030	0.8114	0.8110	0.8108	0.8114
0.2337	4.7300	32500	0.5624	0.8160	0.8164	0.8171	0.8160
0.2296	4.8028	33000	0.5819	0.8078	0.8081	0.8091	0.8078
0.2401	4.8756	33500	0.6108	0.8132	0.8129	0.8132	0.8132
0.2274	4.9483	34000	0.6007	0.8105	0.8103	0.8101	0.8105
0.1746	5.0211	34500	0.6685	0.8141	0.8145	0.8150	0.8141
0.1809	5.0939	35000	0.6588	0.8150	0.8148	0.8151	0.8150
0.174	5.1666	35500	0.7005	0.8087	0.8090	0.8101	0.8087
0.1826	5.2394	36000	0.6659	0.8160	0.8164	0.8179	0.8160
0.1814	5.3122	36500	0.6624	0.8096	0.8097	0.8099	0.8096
0.1844	5.3850	37000	0.6554	0.8196	0.8197	0.8200	0.8196
0.1881	5.4577	37500	0.6446	0.8205	0.8209	0.8216	0.8205
0.1749	5.5305	38000	0.6848	0.8205	0.8201	0.8200	0.8205
0.1802	5.6033	38500	0.6504	0.8196	0.8197	0.8198	0.8196
0.1859	5.6760	39000	0.6666	0.8214	0.8218	0.8229	0.8214
0.1978	5.7488	39500	0.6629	0.8123	0.8128	0.8137	0.8123
0.1926	5.8216	40000	0.6870	0.8087	0.8086	0.8087	0.8087
0.1938	5.8943	40500	0.6402	0.8259	0.8257	0.8257	0.8259
0.1816	5.9671	41000	0.6729	0.8132	0.8135	0.8146	0.8132
0.1398	6.0399	41500	0.7129	0.8196	0.8199	0.8204	0.8196
0.152	6.1126	42000	0.7323	0.8169	0.8176	0.8205	0.8169
0.1526	6.1854	42500	0.7196	0.8105	0.8103	0.8101	0.8105
0.1492	6.2582	43000	0.7415	0.8078	0.8082	0.8093	0.8078
0.1447	6.3310	43500	0.7232	0.8150	0.8150	0.8150	0.8150
0.1616	6.4037	44000	0.7005	0.8069	0.8070	0.8076	0.8069
0.1395	6.4765	44500	0.7319	0.8150	0.8150	0.8150	0.8150
0.1447	6.5493	45000	0.7096	0.8141	0.8143	0.8146	0.8141
0.152	6.6220	45500	0.6835	0.8069	0.8069	0.8070	0.8069
0.1502	6.6948	46000	0.7245	0.8150	0.8156	0.8164	0.8150
0.1525	6.7676	46500	0.7016	0.8178	0.8182	0.8191	0.8178
0.157	6.8403	47000	0.7406	0.8087	0.8088	0.8093	0.8087
0.1477	6.9131	47500	0.7340	0.8132	0.8129	0.8127	0.8132
0.164	6.9859	48000	0.6883	0.8150	0.8148	0.8147	0.8150
0.1176	7.0587	48500	0.7798	0.8123	0.8121	0.8120	0.8123
0.1202	7.1314	49000	0.7973	0.8141	0.8145	0.8151	0.8141
0.1246	7.2042	49500	0.7621	0.8096	0.8095	0.8096	0.8096
0.1273	7.2770	50000	0.7751	0.8114	0.8118	0.8127	0.8114
0.1207	7.3497	50500	0.7591	0.8141	0.8145	0.8150	0.8141
0.1182	7.4225	51000	0.8026	0.8105	0.8105	0.8107	0.8105
0.1276	7.4953	51500	0.7784	0.8078	0.8076	0.8078	0.8078
0.1313	7.5680	52000	0.7864	0.8096	0.8097	0.8098	0.8096
0.1293	7.6408	52500	0.7635	0.8060	0.8065	0.8072	0.8060
0.1124	7.7136	53000	0.7906	0.8060	0.8066	0.8086	0.8060
0.1284	7.7863	53500	0.7494	0.8114	0.8121	0.8132	0.8114
0.1284	7.8591	54000	0.7634	0.8078	0.8083	0.8096	0.8078
0.1245	7.9319	54500	0.7714	0.8051	0.8055	0.8063	0.8051
0.1221	8.0047	55000	0.7907	0.8033	0.8033	0.8037	0.8033
0.1	8.0774	55500	0.8407	0.8069	0.8073	0.8084	0.8069
0.1012	8.1502	56000	0.8541	0.8087	0.8084	0.8082	0.8087
0.1059	8.2230	56500	0.8524	0.8078	0.8080	0.8086	0.8078
0.1009	8.2957	57000	0.8298	0.8069	0.8075	0.8086	0.8069
0.1017	8.3685	57500	0.8563	0.8051	0.8057	0.8076	0.8051
0.1167	8.4413	58000	0.8364	0.8051	0.8054	0.8062	0.8051
0.106	8.5140	58500	0.8377	0.8033	0.8035	0.8040	0.8033
0.1123	8.5868	59000	0.8270	0.8078	0.8083	0.8099	0.8078
0.1149	8.6596	59500	0.8203	0.8078	0.8079	0.8082	0.8078
0.1043	8.7324	60000	0.8346	0.8087	0.8093	0.8103	0.8087
0.1012	8.8051	60500	0.8428	0.8060	0.8060	0.8061	0.8060
0.0987	8.8779	61000	0.8206	0.8033	0.8036	0.8040	0.8033
0.0944	8.9507	61500	0.8461	0.8042	0.8045	0.8059	0.8042
0.0796	9.0234	62000	0.8857	0.8069	0.8073	0.8082	0.8069
0.0886	9.0962	62500	0.8858	0.8096	0.8102	0.8113	0.8096
0.0865	9.1690	63000	0.8939	0.8069	0.8074	0.8083	0.8069
0.082	9.2417	63500	0.9015	0.8087	0.8092	0.8099	0.8087
0.0956	9.3145	64000	0.8834	0.8087	0.8090	0.8098	0.8087
0.0941	9.3873	64500	0.8920	0.8078	0.8080	0.8084	0.8078
0.084	9.4600	65000	0.8873	0.8042	0.8045	0.8051	0.8042
0.0901	9.5328	65500	0.9064	0.8078	0.8082	0.8088	0.8078
0.0895	9.6056	66000	0.8816	0.8096	0.8102	0.8111	0.8096
0.0984	9.6784	66500	0.8674	0.8087	0.8092	0.8100	0.8087
0.0943	9.7511	67000	0.8739	0.8087	0.8092	0.8101	0.8087
0.0833	9.8239	67500	0.8850	0.8087	0.8091	0.8097	0.8087
0.0809	9.8967	68000	0.8744	0.8051	0.8054	0.8058	0.8051
0.0942	9.9694	68500	0.8762	0.8078	0.8082	0.8089	0.8078

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.2