<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
BERT-Tiny fine-tuned on Enron Spam Detection
This model is a fine-tuned version of google/bert_uncased_L-2_H-128_A-2 (aka BERT-Tiny) on an SetFit/enron_spam for Spam Dectection downstream task.
It achieves the following results on the evaluation set:
- Loss: 0.0593
- Precision: 0.9851
- Recall: 0.9871
- Accuracy: 0.986
- F1: 0.9861
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Accuracy | F1 |
|---|---|---|---|---|---|---|---|
| 0.1125 | 1.0 | 1983 | 0.0797 | 0.9839 | 0.9692 | 0.9765 | 0.9765 |
| 0.061 | 2.0 | 3966 | 0.0618 | 0.9822 | 0.9861 | 0.984 | 0.9842 |
| 0.0486 | 3.0 | 5949 | 0.0593 | 0.9851 | 0.9871 | 0.986 | 0.9861 |
| 0.048 | 4.0 | 7932 | 0.0588 | 0.9870 | 0.9821 | 0.9845 | 0.9846 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1
