bert_base_for_whole_train_result_Spam-Ham_farshad_4_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: 0.0583
• Accuracy: 0.9916
• F1: 0.9919

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
0.5984	2.9250	50	0.4198	0.8741	0.8802
0.2434	5.8501	100	0.1013	0.9710	0.9716
0.0539	8.7751	150	0.0462	0.9840	0.9845
0.0238	11.7002	200	0.0472	0.9849	0.9853
0.0128	14.6252	250	0.0381	0.9884	0.9888
0.0079	17.5503	300	0.0509	0.9875	0.9879
0.0055	20.4753	350	0.0586	0.9846	0.9850
0.0045	23.4004	400	0.0581	0.9887	0.9891
0.005	26.3254	450	0.0494	0.9898	0.9902
0.0036	29.2505	500	0.0614	0.9875	0.9879
0.0018	32.1755	550	0.0466	0.9922	0.9925
0.0023	35.1005	600	0.0508	0.9916	0.9919
0.0016	38.0256	650	0.0960	0.9806	0.9810
0.002	40.9506	700	0.0495	0.9913	0.9916
0.0012	43.8757	750	0.0528	0.9925	0.9927
0.0025	46.8007	800	0.0464	0.9925	0.9927
0.0013	49.7258	850	0.0653	0.9878	0.9882
0.001	52.6508	900	0.0482	0.9925	0.9927
0.0017	55.5759	950	0.0515	0.9922	0.9924
0.0007	58.5009	1000	0.0444	0.9936	0.9938
0.0007	61.4260	1050	0.0522	0.9907	0.9910
0.0005	64.3510	1100	0.0802	0.9875	0.9879
0.0007	67.2761	1150	0.0641	0.9898	0.9902
0.0005	70.2011	1200	0.0546	0.9919	0.9922
0.0005	73.1261	1250	0.0558	0.9922	0.9924
0.0004	76.0512	1300	0.0647	0.9910	0.9913
0.0004	78.9762	1350	0.0599	0.9919	0.9922
0.0004	81.9013	1400	0.0704	0.9913	0.9916
0.0004	84.8263	1450	0.0657	0.9910	0.9913
0.0004	87.7514	1500	0.0583	0.9916	0.9919

Framework versions

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