Skip to content

Latest commit

 

History

History
70 lines (52 loc) · 1.95 KB

README.md

File metadata and controls

70 lines (52 loc) · 1.95 KB
Raw

Spam Detection Web Tool

Spam SMS Detection Project implemented using NLP & Transformers. roBerta - a hugging face Transformer model for text classification is used to fine-tune to best suit data to achieve the best results. The model was deployed on the Flask server.

Demo

Alt text

Description

Training SimpleTransformers model on SMS Spam Collection Dataset, resulting in 0.9993 accuracy on the test results.

Getting Started

Dependencies

  • Please refer to .The pyTorch website to install PyTorch
  • Make the appropriate adjustments to enable Cuda on your PC for faster experience
  • Python 3.8.6
  • transformers 4.17.0
  • torch 1.9.0

Installing

  • git clone this repository
  • Create a virtual environment using
Python -m venv .
  • run
pip install -r /path/to/requirements.txt

Executing program

CMD:

  • Inside the folder load CMD
  • run
python application.py

PyCharm

  • load the projects folder and press Run.
  • The website will automatically load.
  • Insert your messages and press Predict.

Training the model yourself

  • Download the "train_model" folder onto your PC
  • Follow the installations command to set up the environment
  • Execute
python training.py
  • At the end of the training, 'outputs' folder will be created. This is your trained model.

Contributors


Asahel Cohen