Number of steps to be used for gradient accumulation. Gradient accumulation refers to a method of collecting gradient for configured number of steps instead of updating the model variables at every step and then applying the update to model variables. This can be used as a tool to overcome smaller batch size limitation. Often also referred in conjunction with 'effective batch size'.

interface TrainingAccumulatedSteps {
    default?: number;
    max?: number;
    min?: number;
}

Properties

Properties

default?: number

The default value.

max?: number

The maximum value.

min?: number

The minimum value.