Accident-Prediction

Accident prediction using Naive Bayes (Machine Learning CED17 Course Project)

Components of the project

• __init__.py : main python script to start the flask application (server)
• naive_bayes.py : python script including import classes which contain implementation of Naive-Bayesian Model
• accident-datasets/ : directory containing raw, sorted datasets obtained from data.gov.in
• cleaning.ipynb : jupyter-notebook containing code and workflow done for cleaning and organising the datasets
• cleaned/ : directory containing pickled files containing useful information obtained from datasets
• test.ipynb : jupyter-notebook containing code workflow for implementing and testing Naive-Bayesian Model
• templates/ : directory containing Jinja templates to be rendered by flask
• static/ : directory containing static files
• setup.py : script to setup flask project environment

Requirements

• Find in requirements.txt

Usage

• navigate to the current directory first
• run on terminal as:

$ gunicorn Accident_Prediction:app

• server running at http://localhost:8000 or http://127.0.0.1:8000

Deployment

http://accident-chance.herokuapp.com/