bert_base_for_whole_train_result_Spam-Ham_farshad_half_1_1

This model is a fine-tuned version of bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0558
• Accuracy: 0.9919
• F1: 0.9921

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 64
• total_train_batch_size: 4096
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.5654	5.8501	50	0.3296	0.9031	0.9002
0.1833	11.7002	100	0.0922	0.9704	0.9708
0.0439	17.5503	150	0.0594	0.9814	0.9818
0.017	23.4004	200	0.0417	0.9896	0.9899
0.0087	29.2505	250	0.0479	0.9881	0.9885
0.007	35.1005	300	0.0745	0.9823	0.9827
0.0048	40.9506	350	0.0767	0.9832	0.9836
0.004	46.8007	400	0.0704	0.9855	0.9859
0.0045	52.6508	450	0.0581	0.9884	0.9887
0.0031	58.5009	500	0.0486	0.9907	0.9910
0.0017	64.3510	550	0.0447	0.9919	0.9922
0.0015	70.2011	600	0.0624	0.9898	0.9902
0.0018	76.0512	650	0.0589	0.9875	0.9879
0.001	81.9013	700	0.0466	0.9939	0.9941
0.0008	87.7514	750	0.0726	0.9878	0.9881
0.0008	93.6015	800	0.0558	0.9919	0.9921

Framework versions

• Transformers 4.40.0
• Pytorch 2.4.1+cu121
• Datasets 3.1.0
• Tokenizers 0.19.1