kbank-finetune-bert2

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

• Loss: 1.4075
• Accuracy: 0.2913
• Precision: 0.2838
• Recall: 0.2913
• F1: 0.2609

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: 16
• eval_batch_size: 16
• 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

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	1.0	15	0.9443	0.5340	0.3892	0.5340	0.4006
No log	2.0	30	0.9709	0.4078	0.3924	0.4078	0.3931
No log	3.0	45	1.1933	0.3495	0.3332	0.3495	0.2510
No log	4.0	60	1.0654	0.3398	0.3309	0.3398	0.3204
No log	5.0	75	1.0364	0.4272	0.3839	0.4272	0.4044
No log	6.0	90	1.2203	0.3301	0.3195	0.3301	0.3023
No log	7.0	105	1.3716	0.2913	0.2580	0.2913	0.2355
No log	8.0	120	1.3235	0.3204	0.3152	0.3204	0.2993
No log	9.0	135	1.3099	0.3592	0.3490	0.3592	0.3471
No log	10.0	150	1.4075	0.2913	0.2838	0.2913	0.2609

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1