Model Registry service

Overview

Model Registry is the service that exposes API to save, fetch and delete machine learning models. Currently, the service is capable to use Mongo DB and file system to manage machine learning models.
The model registry supports the versioning feature that allows to store several versions of the same model. Model tagging feature is coming later

Exposed endpoints

POST / {model_name} {model_version} {file}
Saves file with the name = model_name and version = model_version

GET / {model_name} {model_version}
Returns a model as a file with the name = model_name and version = model_version

DELETE / {model_name} {model_version}
Deletes from a storage the model with the name = model_name and versions = model_version

GET /health_check
health check endpoint

GET /docs
Swagger UI

GET /redoc
Redoc UI

Project Files Structure

src - source code
tests - unit tests
tests-e2e - end-to-end tests

How to contribute

Contributions are always welcomed. There is a lot of ways how you can help to the project.

• Contribute to the unit tests to make it more reliable.
• Contribute to the end-to-end tests to make it more reliable.
• Look for issues with tag "help wanted" and submit pull requests to address them.
• Open an issue to report problems or recommend new features.