bert-finetuned-ner-new

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0935
• Precision: 0.8648
• Recall: 0.8578
• F1: 0.8613
• Accuracy: 0.9777

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use 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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Accuracy	F1	Validation Loss	Precision	Recall
0.2874	1.0	2489	0.9454	0.7421	0.1544	0.6905	0.8019
0.1734	2.0	4978	0.9477	0.7775	0.1471	0.7101	0.8589
0.1436	3.0	7467	0.9517	0.7969	0.1379	0.7424	0.8599
0.1253	4.0	9956	0.9569	0.8123	0.1263	0.7503	0.8854
0.1113	5.0	12445	0.9597	0.8241	0.1212	0.7712	0.8849
0.0987	6.0	14934	0.9617	0.8391	0.1221	0.7947	0.8887
0.0912	7.0	17423	0.9619	0.8382	0.1245	0.7880	0.8953
0.0829	8.0	19912	0.9633	0.8514	0.1275	0.8131	0.8935
0.0761	9.0	22401	0.9623	0.8473	0.1360	0.7997	0.9009
0.0712	10.0	24890	0.9639	0.8588	0.1330	0.8248	0.8957
0.0659	11.0	27379	0.9636	0.8579	0.1337	0.8178	0.9020
0.0617	12.0	29868	0.9657	0.8629	0.1330	0.8300	0.8985
0.0579	13.0	32357	0.9645	0.8613	0.1393	0.8228	0.9034
0.0553	14.0	34846	0.9641	0.8630	0.1402	0.8242	0.9056
0.0513	15.0	37335	0.9665	0.8676	0.1439	0.8373	0.9001
0.0504	16.0	39824	0.9642	0.8633	0.1488	0.8249	0.9055
0.0487	17.0	42313	0.9658	0.8664	0.1452	0.8318	0.9040
0.0469	18.0	44802	0.9667	0.8708	0.1466	0.8419	0.9017
0.0451	19.0	47291	0.9664	0.8714	0.1490	0.8403	0.9048
0.0442	20.0	49780	0.9667	0.8711	0.1492	0.8399	0.9046
0.1019	6.0	54834	0.9697	0.8130	0.0920	0.7831	0.8453
0.0841	7.0	63973	0.9719	0.8211	0.0857	0.7929	0.8514
0.074	8.0	73112	0.9732	0.8300	0.0818	0.8039	0.8579
0.0678	9.0	82251	0.9735	0.8329	0.0846	0.8052	0.8626
0.0626	10.0	91390	0.9747	0.8394	0.0864	0.8147	0.8656
0.059	11.0	100529	0.9760	0.8444	0.0786	0.8276	0.8620
0.0551	12.0	109668	0.9753	0.8436	0.0880	0.8199	0.8687
0.0526	13.0	118807	0.9752	0.8438	0.0870	0.8198	0.8692
0.0497	14.0	127946	0.9758	0.8466	0.0935	0.8246	0.8699
0.0475	15.0	137085	0.9760	0.8496	0.0936	0.8271	0.8734
0.0459	16.0	146224	0.9755	0.8468	0.0973	0.8211	0.8742
0.044	17.0	155363	0.9764	0.8515	0.0945	0.8301	0.8740
0.0429	18.0	164502	0.9767	0.8528	0.0931	0.8348	0.8715
0.0412	19.0	173641	0.9765	0.8513	0.0950	0.8317	0.8719
0.0404	20.0	182780	0.9764	0.8514	0.0968	0.8304	0.8734

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0