Boilerplate

A boilerplate template to wrap your PyTorch machine learning code as a Flask REST API, complete with Dockerfiles, Tensorboard, etc. This template is designed for classification tasks in PyTorch so minor changes are required for regression tasks. Checkout this simple text-classification repository for an example of how this boilerplate template can be leveraged.

Set Up

pip install cookiecutter invoke requests
cookiecutter gh:practicalAI/boilerplate

Steps

1. Place data in src/datasets or have it in S3, etc.
2. Edit the training.json configuration file in src/configs.
3. Define requirements and setup.py for your package in src.
4. Edit application.py, config.py, and wsgi.py in src.
5. Edit endpoints.py and utils.py functions in src/api.
6. Add ML components in src/{{ cookiecutter.service_name }}.
7. Add unit and e2e tests in src/tests.

Directory structure

{{ cookiecutter.service_name }}/
├── src/                                  - source files
|   ├── api/                                - holds all API scripts
|   |   ├── endpoints.py                      - API endpoint definitions
|   |   ├── operations.py                     - endpoint operation
|   |   └── utils.py                          - api utility functions
|   ├── configs/                            - configuration files
|   |   ├── logging.json                      - logger configuration
|   |   ├── training.json                     - training configuration
|   ├── data/                               - directory of datasets
|   ├── experiments/                        - directory of experiments
|   ├── logs/                               - directory of log files
|   |   ├── errors/                           - error log
|   |   ├── info/                             - info log
|   ├── tensorboard/                        - TensorBoard events
|   ├── tests/                              - tests
|   |   ├── e2e/                              - integration tests
|   |   ├── unit/                             - unit tests
|   ├── {{ cookiecutter.package_name }}/    - ML files
|   |   ├── dataset.py                        - dataset generator functions
|   |   ├── eval.py                           - evaluation operations
|   |   ├── load.py                           - load data
|   |   ├── model.py                          - model functions
|   |   ├── predict.py                        - inference operations
|   |   ├── split.py                          - split data
|   |   ├── train.py                          - training operations
|   |   ├── utilities.py                      - utilities
|   |   ├── vectorize.py                      - vectorize data
|   ├── application.py                      - application script
|   ├── config.py                           - application configuration
|   ├── requirements.txt                    - python package requirements
|   ├── setup.py                            - custom package setup
|   ├── wsgi.py                             - application initialization
├── .dockerignore                         - dockerignore file
├── .gitignore                            - gitignore file
├── Dockerfile                            - Dockerfile for the application
├── CODE_OF_CONDUCT.md                    - code of conduct
├── CODEOWNERS                            - code owner assignments
├── LICENSE                               - license description
└── README.md                             - repository readme