gin-config

Gin Config is a lightweight configuration framework for Python that utilizes dependency injection to effortlessly manage function and class parameter configurations, particularly suited for machine learning tasks. Its primary use case involves simplifying configuration management through the `@gin.configurable` decorator, enabling default parameter values to be set via configuration files, thereby enhancing flexibility without requiring extensive boilerplate code. Notable features include support for nested parameters, integration with TensorFlow and PyTorch, and comprehensive handling of variable bindings through a clear syntax.