Modelpublic

autobio/distilbert-scam-classification-fine-tuned-elder

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes9downloads

Model Card

distilbert-scam-classification-fine-tuned-elder

This model is a fine-tuned version of BothBosu/distilbert-scam-classification-v1 on the None dataset. It achieves the following results on the evaluation set:

—Loss: 1.6626
—Accuracy: 0.925
—Precision: 0.8889
—Recall: 0.9412
—F1: 0.9143

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: 0.004
—trainbatchsize: 64
—evalbatchsize: 32
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—training_steps: 200

Training results

Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
6.6667	20	1.8102	0.9	0.8421	0.9412	0.8889
13.3333	40	1.7751	0.9	0.8421	0.9412	0.8889
20.0	60	1.7390	0.9	0.8421	0.9412	0.8889
26.6667	80	1.7140	0.925	0.8889	0.9412	0.9143
33.3333	100	1.7016	0.925	0.8889	0.9412	0.9143
40.0	120	1.6990	0.925	0.8889	0.9412	0.9143
46.6667	140	1.6819	0.925	0.8889	0.9412	0.9143
53.3333	160	1.6602	0.925	0.8889	0.9412	0.9143
60.0	180	1.6626	0.925	0.8889	0.9412	0.9143
66.6667	200	1.6626	0.925	0.8889	0.9412	0.9143

Framework versions

—Transformers 4.40.0
—Pytorch 2.2.1+cu121
—Datasets 2.19.0
—Tokenizers 0.19.1