bert_base_for_whole_train_result_Spam-Ham_farshad_half_2_4

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: 0.0523
• Accuracy: 0.9898
• F1: 0.9902

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
0.6209	5.8501	50	0.4234	0.8962	0.8988
0.2544	11.7002	100	0.0952	0.9722	0.9727
0.0509	17.5503	150	0.0440	0.9872	0.9876
0.0164	23.4004	200	0.0353	0.9910	0.9913
0.0088	29.2505	250	0.0392	0.9910	0.9913
0.0054	35.1005	300	0.0422	0.9910	0.9913
0.0053	40.9506	350	0.0586	0.9872	0.9876
0.0032	46.8007	400	0.0509	0.9890	0.9893
0.0031	52.6508	450	0.0438	0.9910	0.9913
0.0021	58.5009	500	0.0500	0.9916	0.9919
0.0026	64.3510	550	0.0419	0.9922	0.9924
0.0023	70.2011	600	0.0578	0.9887	0.9890
0.0012	76.0512	650	0.0472	0.9910	0.9913
0.0013	81.9013	700	0.0610	0.9898	0.9901
0.001	87.7514	750	0.0659	0.9898	0.9902
0.0016	93.6015	800	0.0523	0.9898	0.9902

Framework versions

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