weibo-model-4tags

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: 1.0245
• Accuracy: 0.7079
• Precision: 0.7101
• Recall: 0.7079
• F1: 0.7081

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.1091	0.6849	50	1.0191	0.5361	0.6449	0.5361	0.4924
0.7439	1.3699	100	0.8837	0.6306	0.6446	0.6306	0.6280
0.7962	2.0548	150	0.8365	0.6615	0.6886	0.6615	0.6567
0.5132	2.7397	200	0.8698	0.6890	0.6977	0.6890	0.6841
0.2886	3.4247	250	0.9056	0.7096	0.7103	0.7096	0.7092
0.1804	4.1096	300	0.9927	0.7045	0.7071	0.7045	0.7027
0.146	4.7945	350	1.0245	0.7079	0.7101	0.7079	0.7081

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1