bert_base_for_whole_train_result_Spam-Ham_farshad_2_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: 0.0467
• Accuracy: 0.9933
• F1: 0.9936

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
0.6311	2.9250	50	0.4131	0.8675	0.8695
0.2324	5.8501	100	0.0975	0.9713	0.9719
0.0534	8.7751	150	0.0385	0.9869	0.9873
0.0218	11.7002	200	0.0464	0.9852	0.9856
0.0121	14.6252	250	0.0463	0.9858	0.9862
0.0078	17.5503	300	0.0367	0.9916	0.9919
0.0082	20.4753	350	0.0346	0.9922	0.9924
0.0043	23.4004	400	0.0463	0.9904	0.9907
0.0037	26.3254	450	0.0576	0.9887	0.9890
0.0039	29.2505	500	0.0392	0.9916	0.9919
0.0021	32.1755	550	0.0453	0.9907	0.9910
0.0017	35.1005	600	0.0443	0.9913	0.9916
0.0025	38.0256	650	0.0488	0.9901	0.9905
0.0012	40.9506	700	0.0510	0.9919	0.9921
0.0017	43.8757	750	0.0522	0.9910	0.9913
0.0011	46.8007	800	0.0446	0.9927	0.9930
0.0016	49.7258	850	0.0373	0.9927	0.9930
0.0014	52.6508	900	0.0448	0.9916	0.9919
0.0009	55.5759	950	0.0460	0.9930	0.9933
0.0009	58.5009	1000	0.0842	0.9846	0.9850
0.0016	61.4260	1050	0.0356	0.9939	0.9941
0.0009	64.3510	1100	0.0390	0.9933	0.9936
0.0007	67.2761	1150	0.0324	0.9945	0.9947
0.0005	70.2011	1200	0.0434	0.9925	0.9927
0.0005	73.1261	1250	0.0498	0.9893	0.9896
0.0005	76.0512	1300	0.0460	0.9930	0.9933
0.0004	78.9762	1350	0.0480	0.9930	0.9933
0.0004	81.9013	1400	0.0477	0.9930	0.9933
0.0004	84.8263	1450	0.0494	0.9930	0.9933
0.0004	87.7514	1500	0.0490	0.9933	0.9936
0.0004	90.6764	1550	0.0497	0.9933	0.9936
0.0004	93.6015	1600	0.0506	0.9933	0.9936
0.0004	96.5265	1650	0.0467	0.9933	0.9936

Framework versions

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