bert-base-chinese-finetuned-ner-split_food

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: 0.0077
• 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: 30
• eval_batch_size: 30
• 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
3.6798	1.0	1	1.6743	0.0
1.8172	2.0	2	0.6580	0.0
0.746	3.0	3	0.4864	0.0
0.4899	4.0	4	0.3927	0.0
0.401	5.0	5	0.2753	0.0
0.2963	6.0	6	0.2160	0.0
0.2452	7.0	7	0.1848	0.5455
0.2188	8.0	8	0.1471	0.7692
0.1775	9.0	9	0.1131	0.7692
0.1469	10.0	10	0.0864	0.8293
0.1145	11.0	11	0.0621	0.9333
0.0881	12.0	12	0.0432	1.0
0.0702	13.0	13	0.0329	1.0
0.0531	14.0	14	0.0268	1.0
0.044	15.0	15	0.0184	1.0
0.0321	16.0	16	0.0129	1.0
0.0255	17.0	17	0.0101	1.0
0.0236	18.0	18	0.0087	1.0
0.0254	19.0	19	0.0080	1.0
0.0185	20.0	20	0.0077	1.0

Framework versions

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