How ali26m/vehicle-trajectory-tracking works
Project type and purpose
• This is a full-stack vehicle-trajectory tracking and prediction system.
File structure analysis
• Top-level directories:
• Data – raw and preprocessed datasets (CSV, JSON text files).
• Machine_Learning_Model – offline model-training notebook, encoders, and serialized model artifacts.
• Route_Tracking – Python scripts to stream and track live vehicle locations.
• Tree_API – a Flask (app.py) microservice exposing the trained models for inference.
• web – frontend and supporting scripts: authentication (auth.py/html), adding cars (add_cars.py), map UIs (main_map.html, predict.html, powerbi.html), plus two Dockerized services for real-time location ingest (get_real_location) and map data (map_docker).
• Configuration files: two Dockerfiles under web/get_real_location and web/map_docker, plus Python requirements.
README insights
• The README only lists the project name, but the structure implies:
• Offline ML pipeline (model training & saving).
• Real-time data ingestion and tracking.
• REST API for model inference.
• Web frontend for user interaction and visualization.
Final advice
• Every distinct logical component—don’t lump the tracker and ingestion logic with the API.
• Be explicit about offline vs. real-time paths so stakeholders see the end-to-end flow at a glance.