How vignesheilik/ddos_project works
This is a small Flask application for classifying uploaded network-traffic CSVs as DDoS or benign. At startup it trains a Gaussian Naive Bayes model from the bundled dataset only when the model artifact is absent. Users open the upload page, submit a CSV, and receive a rendered prediction. The graph reflects the single sampled implementation; the README provides no additional capability details, and template internals were not sampled.
Main entry points: app.py Flask app · ddos_model.pkl Saved model · index.html Upload page · result.html Prediction page
Components 9
Web workflow
- Flask app
app.py - CSV upload handler
app.py - Upload page
templates/index.html - Prediction page
templates/result.html
Classification
- Model training
app.py - Training dataset
test3.csv - Saved model
ddos_model.pkl - Prediction logic
app.py
Other components
- User
Connections 9
- User → Flask app: opens page
app.py - Flask app → Upload page: renders
app.py - User → CSV upload handler: uploads CSV
app.py - CSV upload handler → Prediction logic: processes file
app.py - Prediction logic → Prediction page: renders prediction
app.py - Flask app → Model training: trains if absent
app.py - Model training → Training dataset: reads
app.py - Model training → Saved model: saves
app.py - Prediction logic → Saved model: loads and predicts
app.py