AT Oracle is a two-stage project involving:
-
the creation of a data pipeline to ingest, process, clean, and store historical public transport usage (via Auckland Transport) and weather data
-
implementation of machine learning models to predict future AT public transport usage based on historical data and weather factors
CURRENT:
- Periodic checking of AT's website for new usage data
- Automated collection of xlsx download links
- Idempotent downloading and processing of xlsx sheets
- Cleaning and safe formatting of data for populating SQL database
PLANNED:
- Weather data pipeline
- Automted addition of new information to database
- Training and testing of ML models for prediction
# Create virtual environment (optional but recommended)
python -m venv .venv
source .venv/bin/activate # Linux / macOS
.venv\Scripts\activate # Windows
# Clone project repository
git clone https://github.com/Al-Bee/AT-Oracle.git
cd AT-Oracle
# Install dependencies
pip install -r requirements.txtTO COME...