BERT_test_graident_accumulation

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

• Loss: 1.3780
• Accuracy: 0.6384

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• 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
No log	1.0	188	0.9354	0.5987
No log	2.0	376	0.9827	0.6208
0.7728	3.0	564	1.1462	0.6298
0.7728	4.0	752	1.3019	0.6323
0.7728	5.0	940	1.3780	0.6384

Framework versions

• Transformers 4.36.0
• Pytorch 2.1.1+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0