ditilbert-spamEmail

This model is a fine-tuned version of distilbert-base-uncased on an enron_spam dataset. It achieves the following results on the evaluation set:

• Loss: 0.0462
• Accuracy: 0.9925

Model description

By calling the API, label 0 means ham message while 1 means spam message.

Intended uses & limitations

This model is used for spam email detection powered by distilbert and sequence classification.

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: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0307	1.0	1983	0.0561	0.989
0.007	2.0	3966	0.0462	0.9925

Framework versions

• Transformers 4.34.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1