scb-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: 1.1107
• Accuracy: 0.4038
• Precision: 0.3823
• Recall: 0.4038
• F1: 0.3871

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	1.0576	0.4423	0.1956	0.4423	0.2713
No log	2.0	30	1.0491	0.4327	0.1932	0.4327	0.2672
No log	3.0	45	1.0501	0.3654	0.3072	0.3654	0.2828
No log	4.0	60	1.0507	0.4327	0.3536	0.4327	0.3849
No log	5.0	75	1.0697	0.375	0.3150	0.375	0.3423
No log	6.0	90	1.0865	0.4423	0.3999	0.4423	0.3812
No log	7.0	105	1.0720	0.375	0.3639	0.375	0.3526
No log	8.0	120	1.1018	0.4327	0.4044	0.4327	0.3992
No log	9.0	135	1.1086	0.4327	0.4083	0.4327	0.4107
No log	10.0	150	1.1107	0.4038	0.3823	0.4038	0.3871

Framework versions

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