bert_base_for_whole_train_result_Spam-Ham_farshad_half_4_2

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: 0.0513
• Accuracy: 0.9927
• F1: 0.9930

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
0.6012	5.8501	50	0.4132	0.9008	0.8986
0.2316	11.7002	100	0.0831	0.9771	0.9776
0.0448	17.5503	150	0.0384	0.9878	0.9882
0.0158	23.4004	200	0.0408	0.9884	0.9887
0.0083	29.2505	250	0.0427	0.9896	0.9899
0.0055	35.1005	300	0.0745	0.9852	0.9856
0.0048	40.9506	350	0.0522	0.9875	0.9879
0.0044	46.8007	400	0.0481	0.9890	0.9893
0.0017	52.6508	450	0.0443	0.9916	0.9918
0.0014	58.5009	500	0.0494	0.9922	0.9924
0.0015	64.3510	550	0.0565	0.9896	0.9899
0.0024	70.2011	600	0.0676	0.9872	0.9876
0.0011	76.0512	650	0.0643	0.9896	0.9899
0.0009	81.9013	700	0.0615	0.9910	0.9913
0.0056	87.7514	750	0.0426	0.9927	0.9930
0.0008	93.6015	800	0.0513	0.9927	0.9930

Framework versions

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