Optimizer

Optimizer interface.

Every optimizer inherits from this base interface.

Parameters

• lr

  Type → int | float | Scheduler

Attributes

• learning_rate (float)

  Returns the current learning rate value.

Methods

look_ahead

Updates a weight vector before a prediction is made.

Parameters: w (dict): A dictionary of weight parameters. The weights are modified in-place. Returns: The updated weights.

Parameters

• w — 'dict'

step

Updates a weight vector given a gradient.

Parameters

• w — 'dict | VectorLike'
• g — 'dict | VectorLike'

Returns

dict | VectorLike: The updated weights.