ner-btc

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

• Loss: 0.4142
• Precision: 0.5949
• Recall: 0.6021
• F1: 0.5985
• Accuracy: 0.9297

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 with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	56	0.2800	0.4550	0.4854	0.4697	0.9137
No log	2.0	112	0.2701	0.5275	0.5918	0.5578	0.9250
No log	3.0	168	0.2836	0.5364	0.6072	0.5696	0.9248
No log	4.0	224	0.3103	0.5953	0.5626	0.5785	0.9287
No log	5.0	280	0.3210	0.5794	0.5883	0.5838	0.9284
No log	6.0	336	0.3376	0.5574	0.5832	0.5700	0.9254
No log	7.0	392	0.3717	0.6014	0.5849	0.5930	0.9304
No log	8.0	448	0.3788	0.6017	0.6038	0.6027	0.9297
0.0751	9.0	504	0.3832	0.5972	0.5901	0.5936	0.9296
0.0751	10.0	560	0.3943	0.5686	0.5969	0.5824	0.9265
0.0751	11.0	616	0.3914	0.6042	0.5969	0.6005	0.9306
0.0751	12.0	672	0.4034	0.5892	0.6003	0.5947	0.9286
0.0751	13.0	728	0.4093	0.5963	0.6003	0.5983	0.9297
0.0751	14.0	784	0.4122	0.5973	0.6003	0.5988	0.9294
0.0751	15.0	840	0.4142	0.5949	0.6021	0.5985	0.9297

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1