Pipeline Structure
The evoML pipeline is organized into several key directories and files, each serving a specific purpose in the machine learning workflow:
pipeline_root/
│
├── data/
│ ├── file.json # Metrics for test, train and validation
│ ├── infos.json # Platform specific settings
│ ├── model_artifacts/ # Folder containing model files
│ │ └── model.joblib
│ ├── predict.config.json # Platform specific settings for prediction
│ ├── prepro.joblib # evoML platform data preprocessing code
│ └── sample.csv # Sample dataset
│
├── docs/ # Documentation files
│ ├── CHANGELOG.md
│ ├── model_files.md
│ ├── useful_scripts.md
│ ├── py_scripts_execution.md
│ ├── conda_basics.md
│ ├── jupiter_env.md
│ ├── development.md
│ ├── environment_configuration.md
│ ├── README_API.md
│ └── README_MLFLOW.md
│
├── notebooks/ # Jupyter notebooks for interactive analysis and experimentation.
│ └── notebook.ipynb
│
├── setup/
│ ├── requirements.txt
│ ├── requirements_develop.txt
│ ├── requirements_ipynb.txt
│ ├── requirements_mlflow.txt
│ ├── requirements_prepro.txt
│ ├── requirements_rest.txt
│ └── dist/
│ └── *.whl # Wheel files
│
├── src/ # Source code for the main pipeline
│ ├── pipeline/
│ │ ├── predict.py
│ │ ├── train.py
│ │ ├── save.py
│ │ ├── update.py
│ │ └── evaluate.py
│ └── pipeline_rest/ # REST API related code
│
├── src_mlops/ # MLOps related source code
│
├── tests/ # Unit and integration tests
│
├── deploy/ # Deployment configuration
│ ├── .env
│ └── docker/
│ └── .env
│
├── LICENSE # License file
├── Makefile # Makefile for various commands (for Linux and Mac, no Windows support)
├── pyproject.toml # Python project configuration (for poetry, for Windows)
├── setup.cfg # Configuration file for setup.py
└── README.md # Main README file