Arkadiusz Ostrowski
← Back to projects

Phishing Detection Pipeline

Implements data ingestion (MongoDB), schema-based validation (YAML), preprocessing, model training/selection, and experiment tracking with MLflow (DagsHub). Exposes a /train endpoint and a CSV-based /predict upload that returns an HTML report with predictions.

PythonFastAPIscikit-learnpandasNumPyMongoDB (PyMongo)MLflowDagsHubDockerYAMLJinja2

Demo

Screenshots and a short video are coming soon. In the meantime, feel free to check the repository for implementation details.