distilbert-spamEmail

This model is a second fine-tuned version of tony4194/ditilbert-spamEmail on an SetFit/enron_spam dataset. It achieves the following results on the evaluation set:

• Loss: 0.0621
• Accuracy: 0.9925

Model description

To detect spam messages.

Intended uses & limitations

Maximum paragraph or chunk_text is 512.

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0083	1.0	1983	0.0569	0.9915
0.0012	2.0	3966	0.0620	0.993
0.0003	3.0	5949	0.0621	0.9925

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1