bert-base-chinese-finetuned-ner

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

• Loss: 0.0063
• F1: 1.0

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
3.0712	1.0	3	1.6814	0.0472
1.545	2.0	6	1.1195	0.4993
1.1234	3.0	9	0.7210	0.7259
0.6518	4.0	12	0.4457	0.8595
0.497	5.0	15	0.2754	0.9050
0.2761	6.0	18	0.1742	0.9509
0.2281	7.0	21	0.1053	0.9903
0.1189	8.0	24	0.0642	0.9976
0.1002	9.0	27	0.0416	1.0
0.053	10.0	30	0.0280	1.0
0.0525	11.0	33	0.0206	1.0
0.0412	12.0	36	0.0156	1.0
0.0284	13.0	39	0.0123	1.0
0.0191	14.0	42	0.0101	1.0
0.0227	15.0	45	0.0087	1.0
0.0167	16.0	48	0.0077	1.0
0.0161	17.0	51	0.0071	1.0
0.015	18.0	54	0.0066	1.0
0.0167	19.0	57	0.0064	1.0
0.0121	20.0	60	0.0063	1.0

Framework versions

• Transformers 4.18.0
• Pytorch 1.12.0+cu102
• Datasets 1.18.4
• Tokenizers 0.12.1