bert_base_for_whole_train_result_Spam-Ham2_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: 0.0253
• Accuracy: 0.9965
• F1: 0.9967

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
0.6911	6.8817	50	0.5512	0.862	0.8583
0.2064	13.7634	100	0.0465	0.9865	0.9873
0.0144	20.6452	150	0.0371	0.9905	0.9910
0.0039	27.5269	200	0.0308	0.9945	0.9948
0.0027	34.4086	250	0.0355	0.994	0.9944
0.0013	41.2903	300	0.0446	0.9925	0.9929
0.0005	48.1720	350	0.0381	0.994	0.9944
0.0006	55.0538	400	0.0391	0.994	0.9944
0.0016	61.9355	450	0.0403	0.993	0.9934
0.0003	68.8172	500	0.0350	0.9955	0.9958
0.0004	75.6989	550	0.0361	0.994	0.9944
0.0002	82.5806	600	0.0366	0.995	0.9953
0.0004	89.4624	650	0.0375	0.9945	0.9948
0.0007	96.3441	700	0.0253	0.9965	0.9967

Framework versions

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