Fullstack application for malware detection using CNN

This fullstack web application is written in ReactJS(Frontend) and FastAPI(Backend). The code uses VGG16 and Densenet for classification. You can change the code in backend/classifiers.py to use as many models you want.

Due to size limitations, the CNN models have not been uploaded. You have to train your model before running this application.

Prerequisites

• python3.6+
• Node v10
• yarn
• Tensorflow

Frontend setup

• cd to frontend folder
• run nmp install

Backend setup

• cd to backend folder
• create a new directory called trained_models and copy your saved h5 model here
• run pip install -r requirements.txt

Start the frontend server (within frontend folder):

• npm start

Start the backend server (within backend folder):

• uvicorn app:app --reload

Contributors:

Mohammed Annan Greeshma Elachitaya Mridul Tiwary