“平滑"背后的数学原理是什么?TensorBoard 标量图中的参数? [英] What is the mathematics behind the "smoothing" parameter in TensorBoard's scalar graphs?

问题描述

我认为它是某种移动平均线，但有效范围在 0 到 1 之间.

I presume it is some kind of moving average, but the valid range is between 0 and 1.

推荐答案

它叫做指数移动平均线，下面是代码解释它是如何创建的.

It is called exponential moving average, below is a code explanation how it is created.

假设所有实数标量值都在一个名为scalars的列表中，平滑应用如下:

Assuming all the real scalar values are in a list called scalars the smoothing is applied as follows:

def smooth(scalars: List[float], weight: float) -> List[float]:  # Weight between 0 and 1
    last = scalars[0]  # First value in the plot (first timestep)
    smoothed = list()
    for point in scalars:
        smoothed_val = last * weight + (1 - weight) * point  # Calculate smoothed value
        smoothed.append(smoothed_val)                        # Save it
        last = smoothed_val                                  # Anchor the last smoothed value

    return smoothed

这篇关于“平滑"背后的数学原理是什么?TensorBoard 标量图中的参数?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！