ner-coin

This model is a fine-tuned version of microsoft/Multilingual-MiniLM-L12-H384 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0336
• Precision: 0.9730
• Recall: 0.9789
• F1: 0.9759
• Accuracy: 0.9937

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	26	0.2470	0.7608	0.8733	0.8132	0.9776
No log	2.0	52	0.1881	0.7803	0.8733	0.8242	0.9817
No log	3.0	78	0.1495	0.9629	0.9789	0.9708	0.9929
No log	4.0	104	0.1277	0.9744	0.9759	0.9751	0.9942
No log	5.0	130	0.1107	0.9745	0.9804	0.9774	0.9949
No log	6.0	156	0.1004	0.9688	0.9819	0.9753	0.9940
No log	7.0	182	0.0895	0.9701	0.9804	0.9752	0.9942
No log	8.0	208	0.0808	0.9745	0.9789	0.9767	0.9949
No log	9.0	234	0.0731	0.9744	0.9774	0.9759	0.9947
No log	10.0	260	0.0720	0.9731	0.9819	0.9775	0.9939
No log	11.0	286	0.0662	0.9731	0.9819	0.9775	0.9940
No log	12.0	312	0.0589	0.9804	0.9804	0.9804	0.9955
No log	13.0	338	0.0573	0.9746	0.9819	0.9782	0.9945
No log	14.0	364	0.0520	0.9789	0.9789	0.9789	0.9945
No log	15.0	390	0.0507	0.9803	0.9774	0.9789	0.9947
No log	16.0	416	0.0475	0.9804	0.9789	0.9796	0.9949
No log	17.0	442	0.0461	0.9731	0.9804	0.9767	0.9944
No log	18.0	468	0.0435	0.9773	0.9744	0.9758	0.9947
No log	19.0	494	0.0400	0.9760	0.9819	0.9789	0.9952
0.1028	20.0	520	0.0390	0.9834	0.9819	0.9826	0.9960
0.1028	21.0	546	0.0386	0.9716	0.9804	0.9760	0.9945
0.1028	22.0	572	0.0373	0.9688	0.9834	0.9760	0.9942
0.1028	23.0	598	0.0355	0.9789	0.9804	0.9797	0.9950
0.1028	24.0	624	0.0381	0.9617	0.9834	0.9724	0.9924
0.1028	25.0	650	0.0328	0.9775	0.9819	0.9797	0.9950
0.1028	26.0	676	0.0329	0.9789	0.9804	0.9797	0.9952
0.1028	27.0	702	0.0357	0.9789	0.9804	0.9797	0.9950
0.1028	28.0	728	0.0357	0.9688	0.9849	0.9768	0.9940
0.1028	29.0	754	0.0382	0.9632	0.9864	0.9747	0.9921
0.1028	30.0	780	0.0303	0.9789	0.9819	0.9804	0.9953
0.1028	31.0	806	0.0289	0.9819	0.9819	0.9819	0.9957
0.1028	32.0	832	0.0296	0.9790	0.9834	0.9812	0.9955
0.1028	33.0	858	0.0290	0.9848	0.9789	0.9818	0.9953
0.1028	34.0	884	0.0301	0.9789	0.9819	0.9804	0.9953
0.1028	35.0	910	0.0294	0.9702	0.9834	0.9768	0.9944
0.1028	36.0	936	0.0347	0.9717	0.9834	0.9775	0.9936
0.1028	37.0	962	0.0303	0.9746	0.9819	0.9782	0.9939
0.1028	38.0	988	0.0344	0.9645	0.9849	0.9746	0.9923
0.0209	39.0	1014	0.0300	0.9717	0.9834	0.9775	0.9937
0.0209	40.0	1040	0.0288	0.9789	0.9819	0.9804	0.9950
0.0209	41.0	1066	0.0289	0.9804	0.9819	0.9812	0.9952
0.0209	42.0	1092	0.0296	0.9716	0.9804	0.9760	0.9939
0.0209	43.0	1118	0.0319	0.9659	0.9834	0.9746	0.9928
0.0209	44.0	1144	0.0269	0.9848	0.9759	0.9803	0.9950
0.0209	45.0	1170	0.0259	0.9804	0.9804	0.9804	0.9950
0.0209	46.0	1196	0.0306	0.9716	0.9819	0.9767	0.9939
0.0209	47.0	1222	0.0354	0.9658	0.9789	0.9723	0.9918
0.0209	48.0	1248	0.0280	0.9746	0.9819	0.9782	0.9937
0.0209	49.0	1274	0.0266	0.9833	0.9774	0.9803	0.9955
0.0209	50.0	1300	0.0287	0.9760	0.9819	0.9789	0.9945
0.0209	51.0	1326	0.0280	0.9818	0.9759	0.9788	0.9950
0.0209	52.0	1352	0.0316	0.9787	0.9713	0.9750	0.9937
0.0209	53.0	1378	0.0302	0.9744	0.9774	0.9759	0.9936
0.0209	54.0	1404	0.0309	0.9744	0.9759	0.9751	0.9932
0.0209	55.0	1430	0.0298	0.9818	0.9759	0.9788	0.9947
0.0209	56.0	1456	0.0291	0.9729	0.9744	0.9736	0.9931
0.0209	57.0	1482	0.0287	0.9773	0.9759	0.9766	0.9937
0.0099	58.0	1508	0.0349	0.9687	0.9789	0.9737	0.9921
0.0099	59.0	1534	0.0295	0.9745	0.9789	0.9767	0.9936
0.0099	60.0	1560	0.0306	0.9759	0.9789	0.9774	0.9936
0.0099	61.0	1586	0.0298	0.9775	0.9819	0.9797	0.9944
0.0099	62.0	1612	0.0296	0.9746	0.9819	0.9782	0.9944
0.0099	63.0	1638	0.0282	0.9760	0.9804	0.9782	0.9944
0.0099	64.0	1664	0.0290	0.9804	0.9804	0.9804	0.9949
0.0099	65.0	1690	0.0290	0.9745	0.9789	0.9767	0.9937
0.0099	66.0	1716	0.0277	0.9774	0.9789	0.9781	0.9944
0.0099	67.0	1742	0.0303	0.9745	0.9804	0.9774	0.9942
0.0099	68.0	1768	0.0283	0.9773	0.9759	0.9766	0.9945
0.0099	69.0	1794	0.0301	0.9759	0.9774	0.9766	0.9940
0.0099	70.0	1820	0.0304	0.9745	0.9789	0.9767	0.9940
0.0099	71.0	1846	0.0290	0.9789	0.9774	0.9781	0.9944
0.0099	72.0	1872	0.0346	0.9658	0.9789	0.9723	0.9926
0.0099	73.0	1898	0.0327	0.9687	0.9789	0.9737	0.9932
0.0099	74.0	1924	0.0315	0.9759	0.9789	0.9774	0.9940
0.0099	75.0	1950	0.0305	0.9774	0.9774	0.9774	0.9940
0.0099	76.0	1976	0.0304	0.9759	0.9789	0.9774	0.9942
0.0059	77.0	2002	0.0306	0.9716	0.9789	0.9752	0.9936
0.0059	78.0	2028	0.0304	0.9789	0.9789	0.9789	0.9944
0.0059	79.0	2054	0.0322	0.9687	0.9789	0.9737	0.9932
0.0059	80.0	2080	0.0323	0.9730	0.9789	0.9759	0.9936
0.0059	81.0	2106	0.0314	0.9730	0.9789	0.9759	0.9937
0.0059	82.0	2132	0.0315	0.9730	0.9789	0.9759	0.9937
0.0059	83.0	2158	0.0310	0.9731	0.9804	0.9767	0.9939
0.0059	84.0	2184	0.0318	0.9701	0.9804	0.9752	0.9936
0.0059	85.0	2210	0.0317	0.9745	0.9789	0.9767	0.9939
0.0059	86.0	2236	0.0316	0.9745	0.9789	0.9767	0.9939
0.0059	87.0	2262	0.0318	0.9745	0.9789	0.9767	0.9939
0.0059	88.0	2288	0.0324	0.9730	0.9789	0.9759	0.9937
0.0059	89.0	2314	0.0320	0.9745	0.9789	0.9767	0.9939
0.0059	90.0	2340	0.0336	0.9701	0.9789	0.9745	0.9934
0.0059	91.0	2366	0.0335	0.9730	0.9789	0.9759	0.9937
0.0059	92.0	2392	0.0332	0.9745	0.9789	0.9767	0.9939
0.0059	93.0	2418	0.0334	0.9730	0.9789	0.9759	0.9937
0.0059	94.0	2444	0.0335	0.9716	0.9789	0.9752	0.9936
0.0059	95.0	2470	0.0342	0.9701	0.9789	0.9745	0.9931
0.0059	96.0	2496	0.0338	0.9716	0.9789	0.9752	0.9936
0.0045	97.0	2522	0.0338	0.9730	0.9789	0.9759	0.9937
0.0045	98.0	2548	0.0338	0.9716	0.9789	0.9752	0.9936
0.0045	99.0	2574	0.0336	0.9730	0.9789	0.9759	0.9937
0.0045	100.0	2600	0.0336	0.9730	0.9789	0.9759	0.9937

Framework versions

• Transformers 4.40.2
• Pytorch 2.1.0+cu121
• Datasets 2.14.5
• Tokenizers 0.19.1