Model Card for Text Classification for email-spam detection
This model is based on Text classification using pytorch library. In this model we propose to used a torchtext library for tokenize & vectorize data. This model is used in corporate and industrial area for mail detection. It is used three label like job, enquiry and spam. It achieve the following results on the evalution set:
- accuracy : 0.866
model architecture for text classification :
Label for text classification:
- Enquiry
- Job
- Spam
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.01
- train_batch_size: 64
- step_size: 10
- optimizer: Adam
- lr_scheduler_type: StepLR
- lr_scheduler.StepLR:(optimizer,step_size=10,gamma=0.1)
- num_epochs: 10
Framework versions
- Pytorch 2.0.1+cu118
- torchtext 0.15.2+cpu
@ModelCard{
author = {Nehul Agrawal and
Rahul parihar},
title = {Text classification},
year = {2023}
}
Inference API (serverless) does not yet support Pytorch models for this pipeline type.
Evaluation results
- precisionself-reported0.866