Ner-vibert4news-base-cased

This model is a fine-tuned version of NlpHUST/electra-base-vn on the hts98/UIT dataset. It achieves the following results on the evaluation set:

Loss: 2.2270
Precision: 0.6232
Recall: 0.6731
F1: 0.6472
Accuracy: 0.7938

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: 3e-05
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 120.0

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	487	0.7563	0.4074	0.5740	0.4765	0.7624
1.1928	2.0	974	0.6945	0.4871	0.6226	0.5466	0.7788
0.6432	3.0	1461	0.7379	0.5035	0.6348	0.5616	0.7798
0.4783	4.0	1948	0.7334	0.5022	0.6396	0.5626	0.7849
0.3609	5.0	2435	0.8053	0.5322	0.6535	0.5866	0.7819
0.2735	6.0	2922	0.8289	0.5283	0.6401	0.5789	0.7850
0.2243	7.0	3409	0.9323	0.5463	0.6396	0.5892	0.7812
0.1753	8.0	3896	0.9913	0.5343	0.6547	0.5884	0.7798
0.147	9.0	4383	1.0703	0.5336	0.6535	0.5875	0.7738
0.1204	10.0	4870	1.0797	0.5470	0.6357	0.5880	0.7853
0.1046	11.0	5357	1.1283	0.5346	0.6527	0.5878	0.7804
0.0839	12.0	5844	1.1800	0.5472	0.6516	0.5949	0.7841
0.0702	13.0	6331	1.2153	0.5662	0.6600	0.6095	0.7898
0.0631	14.0	6818	1.2912	0.5451	0.6443	0.5906	0.7824
0.0508	15.0	7305	1.3288	0.5665	0.6505	0.6056	0.7855
0.0463	16.0	7792	1.4110	0.5716	0.6485	0.6076	0.7857
0.0387	17.0	8279	1.4022	0.5641	0.6563	0.6067	0.7860
0.0352	18.0	8766	1.4306	0.5540	0.6555	0.6005	0.7892
0.031	19.0	9253	1.4502	0.5659	0.6482	0.6043	0.7931
0.0286	20.0	9740	1.5111	0.5469	0.6572	0.5970	0.7867
0.0262	21.0	10227	1.6086	0.5745	0.6468	0.6085	0.7826
0.0212	22.0	10714	1.6134	0.5790	0.6622	0.6178	0.7845
0.0188	23.0	11201	1.6032	0.5662	0.6533	0.6066	0.7831
0.0179	24.0	11688	1.6524	0.5649	0.6561	0.6071	0.7821
0.0156	25.0	12175	1.6468	0.5643	0.6614	0.6090	0.7832
0.0165	26.0	12662	1.6683	0.5753	0.6524	0.6115	0.7857
0.0141	27.0	13149	1.6882	0.5740	0.6499	0.6096	0.7868
0.013	28.0	13636	1.7041	0.5761	0.6443	0.6083	0.7892
0.0125	29.0	14123	1.7665	0.5803	0.6549	0.6153	0.7868
0.0113	30.0	14610	1.7239	0.5814	0.6538	0.6155	0.7935
0.0115	31.0	15097	1.8083	0.5721	0.6535	0.6101	0.7848
0.0121	32.0	15584	1.7592	0.5660	0.6628	0.6106	0.7925
0.0105	33.0	16071	1.7803	0.5799	0.6572	0.6161	0.7882
0.0089	34.0	16558	1.8192	0.5786	0.6513	0.6128	0.7871
0.0107	35.0	17045	1.8329	0.5668	0.6597	0.6097	0.7860
0.011	36.0	17532	1.8010	0.5714	0.6547	0.6102	0.7834
0.0087	37.0	18019	1.8314	0.5906	0.6544	0.6208	0.7898
0.0075	38.0	18506	1.8428	0.5912	0.6577	0.6227	0.7913
0.0075	39.0	18993	1.8757	0.5816	0.6678	0.6217	0.7893
0.0079	40.0	19480	1.8514	0.5897	0.6586	0.6223	0.7897
0.0086	41.0	19967	1.8783	0.5878	0.6655	0.6242	0.7897
0.0075	42.0	20454	1.8177	0.5868	0.6644	0.6232	0.7951
0.0071	43.0	20941	1.8850	0.6038	0.6650	0.6329	0.7940
0.0068	44.0	21428	1.9210	0.5996	0.6661	0.6311	0.7918
0.006	45.0	21915	1.9289	0.5892	0.6630	0.6239	0.7913
0.0077	46.0	22402	1.9011	0.5876	0.6602	0.6218	0.7938
0.0047	47.0	22889	1.9092	0.5856	0.6630	0.6219	0.7934
0.0073	48.0	23376	1.9654	0.5886	0.6639	0.6240	0.7885
0.0058	49.0	23863	1.9483	0.5809	0.6639	0.6196	0.7884
0.0081	50.0	24350	1.9434	0.5995	0.6566	0.6268	0.7899
0.0063	51.0	24837	1.9490	0.5938	0.6639	0.6269	0.7903
0.0052	52.0	25324	1.9654	0.6072	0.6552	0.6303	0.7879
0.007	53.0	25811	1.9699	0.5967	0.6591	0.6263	0.7880
0.0047	54.0	26298	1.9713	0.5967	0.6614	0.6274	0.7909
0.0041	55.0	26785	1.9534	0.5909	0.6630	0.6249	0.7895
0.0042	56.0	27272	1.9982	0.6028	0.6630	0.6315	0.7941
0.0045	57.0	27759	1.9968	0.6058	0.6544	0.6292	0.7921
0.0045	58.0	28246	1.9851	0.6039	0.6580	0.6298	0.7905
0.0039	59.0	28733	2.0431	0.6067	0.6653	0.6346	0.7891
0.0048	60.0	29220	2.0036	0.5953	0.6494	0.6212	0.7878
0.004	61.0	29707	1.9971	0.6022	0.6669	0.6329	0.7914
0.0032	62.0	30194	2.0073	0.6025	0.6605	0.6302	0.7912
0.0033	63.0	30681	2.0134	0.5962	0.6608	0.6269	0.7918
0.0035	64.0	31168	2.0015	0.5981	0.6619	0.6284	0.7937
0.0032	65.0	31655	1.9974	0.5905	0.6650	0.6255	0.7940
0.0036	66.0	32142	2.0523	0.5935	0.6672	0.6282	0.7892
0.0027	67.0	32629	2.0683	0.6010	0.6695	0.6334	0.7901
0.0039	68.0	33116	2.1081	0.5919	0.6608	0.6245	0.7876
0.0027	69.0	33603	2.0555	0.5973	0.6655	0.6296	0.7923
0.003	70.0	34090	2.1007	0.5912	0.6614	0.6243	0.7880
0.0023	71.0	34577	2.0916	0.6085	0.6709	0.6382	0.7937
0.0016	72.0	35064	2.1564	0.5940	0.6600	0.6252	0.7908
0.0028	73.0	35551	2.1620	0.5947	0.6633	0.6272	0.7863
0.0028	74.0	36038	2.1390	0.5991	0.6683	0.6318	0.7892
0.0025	75.0	36525	2.1204	0.6026	0.6681	0.6337	0.7925
0.0026	76.0	37012	2.1700	0.6011	0.6614	0.6298	0.7884
0.0026	77.0	37499	2.1478	0.5994	0.6639	0.6300	0.7924
0.0022	78.0	37986	2.1547	0.5954	0.6650	0.6282	0.7879
0.0026	79.0	38473	2.1489	0.5851	0.6686	0.6241	0.7879
0.0017	80.0	38960	2.1789	0.5903	0.6706	0.6279	0.7870
0.0016	81.0	39447	2.1882	0.6026	0.6639	0.6318	0.7877
0.0014	82.0	39934	2.1825	0.6015	0.6711	0.6344	0.7880
0.0019	83.0	40421	2.1753	0.6013	0.6661	0.6321	0.7903
0.0014	84.0	40908	2.1887	0.6001	0.6661	0.6314	0.7911
0.0011	85.0	41395	2.1974	0.6055	0.6667	0.6346	0.7913
0.0019	86.0	41882	2.1918	0.6025	0.6678	0.6335	0.7913
0.0014	87.0	42369	2.1962	0.6133	0.6588	0.6353	0.7901
0.0019	88.0	42856	2.1974	0.5953	0.6628	0.6272	0.7902
0.0009	89.0	43343	2.1818	0.6002	0.6667	0.6317	0.7918
0.0016	90.0	43830	2.2059	0.6140	0.6706	0.6410	0.7945
0.0013	91.0	44317	2.2013	0.6086	0.6720	0.6387	0.7922
0.001	92.0	44804	2.1723	0.6084	0.6689	0.6372	0.7945
0.0012	93.0	45291	2.1967	0.6104	0.6706	0.6391	0.7966
0.0023	94.0	45778	2.2024	0.6157	0.6695	0.6414	0.7939
0.0012	95.0	46265	2.2250	0.6097	0.6748	0.6406	0.7929
0.0015	96.0	46752	2.1938	0.6204	0.6734	0.6458	0.7914
0.0012	97.0	47239	2.1854	0.6012	0.6801	0.6382	0.7897
0.0008	98.0	47726	2.2005	0.6199	0.6734	0.6455	0.7930
0.0008	99.0	48213	2.1999	0.6088	0.6731	0.6394	0.7896
0.0011	100.0	48700	2.2228	0.6086	0.6695	0.6376	0.7931
0.0006	101.0	49187	2.2300	0.6110	0.6784	0.6429	0.7925
0.001	102.0	49674	2.2194	0.6059	0.6748	0.6385	0.7917
0.0007	103.0	50161	2.2048	0.6131	0.6742	0.6422	0.7947
0.0003	104.0	50648	2.2270	0.6232	0.6731	0.6472	0.7938
0.0008	105.0	51135	2.2284	0.6184	0.6742	0.6451	0.7952
0.0005	106.0	51622	2.2278	0.6080	0.6742	0.6394	0.7921
0.0004	107.0	52109	2.2571	0.6157	0.6759	0.6444	0.7926
0.0006	108.0	52596	2.2562	0.6069	0.6723	0.6379	0.7925
0.0005	109.0	53083	2.2255	0.6172	0.6717	0.6433	0.7950
0.0006	110.0	53570	2.2429	0.6104	0.6759	0.6415	0.7931
0.0004	111.0	54057	2.2416	0.6123	0.6742	0.6418	0.7927
0.0004	112.0	54544	2.2629	0.6123	0.6689	0.6394	0.7939
0.0004	113.0	55031	2.2645	0.6136	0.6748	0.6427	0.7932
0.0003	114.0	55518	2.2761	0.6208	0.6736	0.6461	0.7944
0.0004	115.0	56005	2.2684	0.6159	0.6745	0.6438	0.7937
0.0004	116.0	56492	2.2741	0.6160	0.6736	0.6436	0.7926
0.0003	117.0	56979	2.2576	0.6160	0.6736	0.6436	0.7939
0.001	118.0	57466	2.2543	0.6157	0.6731	0.6431	0.7943
0.0003	119.0	57953	2.2541	0.6163	0.6739	0.6438	0.7947
0.0005	120.0	58440	2.2552	0.6164	0.6728	0.6434	0.7944

Framework versions

Transformers 4.32.0.dev0
Pytorch 2.1.0+cu121
Datasets 3.1.0
Tokenizers 0.13.3

hts98/electra-ner

Ner-vibert4news-base-cased

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种