How itskhateeb/mlops-project works
Project type and purpose
• This is an end-to-end MLOps pipeline (a “full-stack” machine learning application).
• It covers data ingestion, preprocessing, model training/evaluation, artifact storage, deployment via Docker, and monitoring/versioning using cloud services (AWS S3, MongoDB).
File structure
• Top-level:
• app.py / demo.py / notebook/ – entry-points (Flask app / demo script / Jupyter demo)
• Dockerfile / .dockerignore / requirements.txt / setup.py / pyproject.toml – containerization & packaging
• config/ – YAML configs for schemas & models
• static/ & templates/ – web UI assets
• src/ holds all Python code, subdivided into:
• src/components – core ML pipeline stages (data_ingestion, transformation, validation, training, evaluation, model_pusher)
• src/pipline – orchestrators (training_pipeline.py, prediction_pipeline.py)
• src/cloud_storage – AWS S3 adapter
• src/configuration – AWS & MongoDB connection setup
• src/data_access – abstractions for data I/O
• src/entity – typed entities (artifacts, configs, estimators)
• src/logger / src/exception / src/utils / src/constants – support modules
README
• Confirms the CI/CD, Dockerization, deployment, monitoring, versioning aspects.