ner_based_bert-base-chinese-only-phone1

This model is a fine-tuned version of google-bert/bert-base-chinese on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0202
• Precision: 0.9861
• Recall: 0.9861
• F1: 0.9861
• Accuracy: 0.9969

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: 128
• eval_batch_size: 128
• 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: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	6	0.6782	0.3163	0.4306	0.3647	0.7385
No log	2.0	12	0.2690	0.2840	0.3194	0.3007	0.9323
No log	3.0	18	0.1381	0.3721	0.4444	0.4051	0.9417
No log	4.0	24	0.0900	0.6216	0.6389	0.6301	0.9688
No log	5.0	30	0.0588	0.9067	0.9444	0.9252	0.9896
No log	6.0	36	0.0412	0.9189	0.9444	0.9315	0.9906
No log	7.0	42	0.0298	0.9324	0.9583	0.9452	0.9917
No log	8.0	48	0.0208	0.9589	0.9722	0.9655	0.9948
No log	9.0	54	0.0223	0.9726	0.9861	0.9793	0.9958
No log	10.0	60	0.0202	0.9861	0.9861	0.9861	0.9969

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1