bert_base_for_whole_train_result_Spam-Ham_farshad_half_2_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: 0.0646
• Accuracy: 0.9907
• F1: 0.9910

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
0.6201	5.8501	50	0.4069	0.8947	0.8909
0.2195	11.7002	100	0.0824	0.9739	0.9743
0.0427	17.5503	150	0.0741	0.9756	0.9760
0.0161	23.4004	200	0.0413	0.9896	0.9899
0.0084	29.2505	250	0.0548	0.9867	0.9870
0.006	35.1005	300	0.0691	0.9855	0.9859
0.0043	40.9506	350	0.0651	0.9867	0.9870
0.0034	46.8007	400	0.0622	0.9887	0.9890
0.0042	52.6508	450	0.0451	0.9910	0.9913
0.0015	58.5009	500	0.0694	0.9875	0.9879
0.0012	64.3510	550	0.0817	0.9855	0.9859
0.0019	70.2011	600	0.0930	0.9829	0.9833
0.0016	76.0512	650	0.0737	0.9881	0.9884
0.0009	81.9013	700	0.0548	0.9916	0.9919
0.0012	87.7514	750	0.0645	0.9907	0.9910
0.0009	93.6015	800	0.0646	0.9907	0.9910

Framework versions

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