bert-finetuned-ner

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

• Loss: 0.0616
• Precision: 0.9319
• Recall: 0.9488
• F1: 0.9403
• Accuracy: 0.9856

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: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	0.1503	264	0.1354	0.7782	0.8544	0.8145	0.9625
0.2679	0.3007	528	0.0971	0.8526	0.9005	0.8759	0.9736
0.2679	0.4510	792	0.0887	0.8900	0.9222	0.9059	0.9781
0.105	0.6014	1056	0.0809	0.9094	0.9278	0.9185	0.9804
0.105	0.7517	1320	0.0714	0.9137	0.9342	0.9239	0.9812
0.0748	0.9021	1584	0.0645	0.9181	0.9377	0.9278	0.9836
0.0748	1.0524	1848	0.0735	0.9173	0.9392	0.9282	0.9825
0.0634	1.2027	2112	0.0692	0.9129	0.9389	0.9257	0.9826
0.0634	1.3531	2376	0.0691	0.9297	0.9478	0.9387	0.9851
0.0428	1.5034	2640	0.0660	0.9229	0.9448	0.9337	0.9844
0.0428	1.6538	2904	0.0602	0.9292	0.9450	0.9370	0.9855
0.0448	1.8041	3168	0.0603	0.9165	0.9461	0.9311	0.9844
0.0448	1.9544	3432	0.0636	0.9311	0.9458	0.9384	0.9848
0.0364	2.1048	3696	0.0686	0.9305	0.9461	0.9383	0.9853
0.0364	2.2551	3960	0.0632	0.9338	0.9497	0.9417	0.9857
0.0211	2.4055	4224	0.0644	0.9284	0.9450	0.9366	0.9845
0.0211	2.5558	4488	0.0628	0.9331	0.9458	0.9394	0.9846
0.0209	2.7062	4752	0.0602	0.9287	0.9473	0.9379	0.9853
0.0221	2.8565	5016	0.0616	0.9319	0.9488	0.9403	0.9856

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1