bert-finetuned-ner

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

• Loss: 0.5677
• Precision: 0.3452
• Recall: 0.5421
• F1: 0.4218
• Accuracy: 0.8688

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	13	0.5728	0.2077	0.3551	0.2621	0.8199
No log	2.0	26	0.5687	0.2889	0.3645	0.3223	0.8312
No log	3.0	39	0.5447	0.2857	0.4486	0.3491	0.8425
No log	4.0	52	0.5509	0.2881	0.4766	0.3592	0.8489
No log	5.0	65	0.5751	0.3121	0.4579	0.3712	0.8511
No log	6.0	78	0.5358	0.3851	0.5794	0.4627	0.8667
No log	7.0	91	0.5484	0.3491	0.5514	0.4275	0.8645
No log	8.0	104	0.5671	0.3580	0.5421	0.4312	0.8672
No log	9.0	117	0.5666	0.3494	0.5421	0.4249	0.8688
No log	10.0	130	0.5677	0.3452	0.5421	0.4218	0.8688

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3