bert-NER-20

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

• Loss: 1.2574
• Precision: 0.0345
• Recall: 0.0148
• F1: 0.0207
• Accuracy: 0.7965

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: 16
• eval_batch_size: 16
• 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	4	1.6763	0.0170	0.0353	0.0229	0.6106
No log	2.0	8	1.3559	0.0326	0.0194	0.0243	0.7806
No log	3.0	12	1.2574	0.0345	0.0148	0.0207	0.7965

Framework versions

• Transformers 4.51.1
• Pytorch 2.5.1+cu124
• Datasets 3.5.0
• Tokenizers 0.21.0