bert-base-cased-finetuned-ner

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

• Loss: 0.2379
• Precision: 0.7741
• Recall: 0.8457
• F1: 0.8083
• Accuracy: 0.9688

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	77	0.3109	0.4607	0.3926	0.4239	0.9271
No log	2.0	154	0.2183	0.5270	0.6330	0.5752	0.9425
No log	3.0	231	0.1719	0.6398	0.7106	0.6734	0.9557
No log	4.0	308	0.1702	0.6518	0.7447	0.6951	0.9544
No log	5.0	385	0.1636	0.7006	0.7766	0.7366	0.9597
No log	6.0	462	0.1784	0.6795	0.8053	0.7371	0.9597
0.1868	7.0	539	0.1870	0.7821	0.7713	0.7766	0.9671
0.1868	8.0	616	0.1796	0.7259	0.8340	0.7762	0.9645
0.1868	9.0	693	0.1771	0.7896	0.8106	0.8000	0.9682
0.1868	10.0	770	0.1892	0.7364	0.8351	0.7827	0.9647
0.1868	11.0	847	0.1949	0.7627	0.8277	0.7939	0.9668
0.1868	12.0	924	0.1951	0.7734	0.8277	0.7996	0.9673
0.0131	13.0	1001	0.2009	0.7592	0.8319	0.7939	0.9666
0.0131	14.0	1078	0.2057	0.7623	0.8426	0.8004	0.9671
0.0131	15.0	1155	0.2067	0.7920	0.8266	0.8090	0.9687
0.0131	16.0	1232	0.2090	0.7301	0.8489	0.7850	0.9638
0.0131	17.0	1309	0.2103	0.7730	0.8330	0.8018	0.9673
0.0131	18.0	1386	0.2118	0.7823	0.8255	0.8033	0.9673
0.0131	19.0	1463	0.2202	0.7436	0.8606	0.7978	0.9654
0.0038	20.0	1540	0.2104	0.7796	0.8394	0.8084	0.9683
0.0038	21.0	1617	0.2168	0.7767	0.8362	0.8053	0.9682
0.0038	22.0	1694	0.2198	0.7713	0.8394	0.8039	0.9686
0.0038	23.0	1771	0.2210	0.7637	0.8457	0.8026	0.9678
0.0038	24.0	1848	0.2215	0.7700	0.8479	0.8071	0.9688
0.0038	25.0	1925	0.2222	0.7656	0.8479	0.8046	0.9684
0.0021	26.0	2002	0.2203	0.7756	0.8457	0.8092	0.9693
0.0021	27.0	2079	0.2249	0.7780	0.85	0.8124	0.9689
0.0021	28.0	2156	0.2315	0.7660	0.8532	0.8072	0.9678
0.0021	29.0	2233	0.2335	0.7740	0.8415	0.8063	0.9685
0.0021	30.0	2310	0.2340	0.7803	0.8351	0.8068	0.9687
0.0021	31.0	2387	0.2340	0.7747	0.8415	0.8067	0.9683
0.0021	32.0	2464	0.2364	0.7728	0.8468	0.8081	0.9683
0.001	33.0	2541	0.2419	0.7645	0.8532	0.8064	0.9678
0.001	34.0	2618	0.2378	0.7750	0.8426	0.8073	0.9683
0.001	35.0	2695	0.2398	0.7692	0.8543	0.8095	0.9683
0.001	36.0	2772	0.2371	0.7743	0.8468	0.8089	0.9688
0.001	37.0	2849	0.2374	0.7711	0.8457	0.8067	0.9686
0.001	38.0	2926	0.2380	0.7715	0.8479	0.8079	0.9686
0.0008	39.0	3003	0.2379	0.7733	0.8457	0.8079	0.9687
0.0008	40.0	3080	0.2379	0.7741	0.8457	0.8083	0.9688

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1