bert_base_for_whole_train_result_Spam-Ham4_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: 0.0297
• Accuracy: 0.995
• F1: 0.9953

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
0.7353	6.8817	50	0.4382	0.8525	0.8511
0.2097	13.7634	100	0.0712	0.982	0.9830
0.0223	20.6452	150	0.0253	0.994	0.9944
0.0042	27.5269	200	0.0379	0.9915	0.9920
0.0018	34.4086	250	0.0323	0.995	0.9953
0.0008	41.2903	300	0.0350	0.994	0.9944
0.0038	48.1720	350	0.0344	0.993	0.9934
0.0014	55.0538	400	0.0310	0.995	0.9953
0.0003	61.9355	450	0.0314	0.9955	0.9958
0.0001	68.8172	500	0.0307	0.9945	0.9948
0.0001	75.6989	550	0.0380	0.994	0.9944
0.0001	82.5806	600	0.0375	0.9945	0.9948
0.0024	89.4624	650	0.0267	0.995	0.9953
0.0003	96.3441	700	0.0297	0.995	0.9953

Framework versions

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