bert-finetuned-ner4

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: 0.1224
• Precision: 0.7984
• Recall: 0.8884
• F1: 0.8410
• Accuracy: 0.9619

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: 8
• eval_batch_size: 8
• 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	Validation Loss	Precision	Recall	F1	Accuracy
0.2804	1.0	2489	0.1538	0.6891	0.8019	0.7413	0.9449
0.1729	2.0	4978	0.1465	0.7065	0.8554	0.7738	0.9470
0.1425	3.0	7467	0.1356	0.7488	0.8644	0.8024	0.9534
0.1244	4.0	9956	0.1270	0.7522	0.8800	0.8111	0.9564
0.1109	5.0	12445	0.1247	0.7649	0.8867	0.8213	0.9588
0.099	6.0	14934	0.1224	0.7984	0.8884	0.8410	0.9619
0.091	7.0	17423	0.1259	0.7906	0.8929	0.8386	0.9618
0.0829	8.0	19912	0.1271	0.8128	0.8905	0.8499	0.9637
0.0756	9.0	22401	0.1300	0.8061	0.8971	0.8492	0.9632
0.0708	10.0	24890	0.1302	0.8233	0.8917	0.8562	0.9635
0.0654	11.0	27379	0.1306	0.8163	0.8944	0.8536	0.9640
0.0612	12.0	29868	0.1310	0.8354	0.8953	0.8643	0.9655
0.0579	13.0	32357	0.1400	0.8173	0.9028	0.8580	0.9631
0.0556	14.0	34846	0.1386	0.8214	0.9019	0.8598	0.9636
0.0512	15.0	37335	0.1451	0.8339	0.9017	0.8665	0.9656
0.0506	16.0	39824	0.1504	0.8179	0.9056	0.8595	0.9629
0.0485	17.0	42313	0.1491	0.8292	0.9034	0.8647	0.9645
0.0464	18.0	44802	0.1450	0.8406	0.9005	0.8695	0.9662
0.0451	19.0	47291	0.1504	0.8336	0.9051	0.8679	0.9649
0.0442	20.0	49780	0.1495	0.8370	0.9033	0.8689	0.9654

Framework versions

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