Skip to content

InverseScaling

Reduces the learning rate using a power schedule.

Assuming an initial learning rate \(\eta\), the learning rate at step \(t\) is:

\[\\frac{eta}{(t + 1) ^ p}\]

where \(p\) is a user-defined parameter.

Parameters

  • learning_rate

    Typefloat

  • power

    Default0.5

Methods

get

Returns the learning rate at a given iteration.

Parameters

  • t'int'