results

This model is a fine-tuned version of bert-base-cased on the conll2003 dataset. It achieves the following results on the evaluation set:

• Loss: 0.0635
• Precision: 0.9307
• Recall: 0.9497
• F1: 0.9401
• Accuracy: 0.9864

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2313	0.2847	500	0.1403	0.8444	0.8696	0.8568	0.9626
0.1088	0.5695	1000	0.0887	0.8717	0.9098	0.8903	0.9765
0.1211	0.8542	1500	0.0846	0.9076	0.9238	0.9156	0.9784
0.0503	1.1390	2000	0.0753	0.9101	0.9354	0.9226	0.9814
0.0493	1.4237	2500	0.0630	0.9170	0.9421	0.9294	0.9833
0.0624	1.7084	3000	0.0705	0.9277	0.9366	0.9321	0.9837
0.0313	1.9932	3500	0.0675	0.9270	0.9426	0.9347	0.9843
0.0335	2.2779	4000	0.0661	0.9284	0.9492	0.9387	0.9857
0.0098	2.5626	4500	0.0693	0.9347	0.9473	0.9410	0.9849
0.0099	2.8474	5000	0.0635	0.9307	0.9497	0.9401	0.9864

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu118
• Datasets 2.19.2
• Tokenizers 0.19.1