bert_base_for_whole_train_result_Spam-Ham_farshad_1_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: 0.0533
• Accuracy: 0.9919
• F1: 0.9922

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
0.5975	2.9250	50	0.3615	0.8910	0.8942
0.186	5.8501	100	0.0843	0.9719	0.9724
0.0512	8.7751	150	0.0875	0.9695	0.9699
0.0224	11.7002	200	0.0550	0.9829	0.9833
0.0123	14.6252	250	0.0736	0.9809	0.9812
0.0085	17.5503	300	0.0605	0.9852	0.9856
0.0071	20.4753	350	0.0452	0.9890	0.9893
0.0044	23.4004	400	0.0473	0.9901	0.9905
0.004	26.3254	450	0.0738	0.9838	0.9841
0.0031	29.2505	500	0.0545	0.9881	0.9885
0.0021	32.1755	550	0.0609	0.9884	0.9887
0.0021	35.1005	600	0.0485	0.9910	0.9913
0.003	38.0256	650	0.0517	0.9893	0.9896
0.0013	40.9506	700	0.0704	0.9881	0.9884
0.0017	43.8757	750	0.0453	0.9901	0.9905
0.0013	46.8007	800	0.0420	0.9939	0.9941
0.0012	49.7258	850	0.0547	0.9893	0.9896
0.001	52.6508	900	0.0496	0.9904	0.9907
0.0009	55.5759	950	0.0559	0.9913	0.9916
0.0018	58.5009	1000	0.0398	0.9925	0.9927
0.0017	61.4260	1050	0.0535	0.9913	0.9916
0.0006	64.3510	1100	0.0514	0.9919	0.9921
0.0005	67.2761	1150	0.0479	0.9930	0.9933
0.0007	70.2011	1200	0.0513	0.9907	0.9910
0.0008	73.1261	1250	0.0524	0.9919	0.9921
0.0004	76.0512	1300	0.0533	0.9919	0.9922

Framework versions

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