openfl.utilities.optimizers.numpy.base_optimizer.Optimizer

class openfl.utilities.optimizers.numpy.base_optimizer.Optimizer

Bases: ABC

Base abstract optimizer class.

This class serves as a base class for all optimizers. It defines the basic structure that all derived optimizer classes should follow. It includes an abstract method step that must be implemented by any concrete optimizer class.

Methods

step	Perform a single step for parameter update.

abstract step(gradients: Dict[str, ndarray]) → None

Perform a single step for parameter update.

This method should be overridden by all subclasses to implement the specific optimization algorithm.

Parameters:

gradients (dict) – Partial derivatives with respect to optimized parameters.