tf.reduce_mean()用法总结

概述

tf.reduce_mean()用于计算tensor(张量)沿着指定的数轴(即tensor的某一维度)上的平均值，用作降维或者计算tensor的平均值。

用法

tf.reduce_mean(input_tensor, axis=None, keepdims=False, name=None)

keep_dims为旧版写法
1.input_tensor：输入待降维的tensor
2.axis：指定的轴，默认为计算所有元素的均值
3.keepdims：是否降低维度，默认值为False。(当为True时，输出结果保持输入tensor的原状；当为False时，输出结果将会降低维度)
4.name：操作的名称

例子

import tensorflow as tf

X = np.array([[0, 1, 2], [3, 4, 5]])
Y = tf.cast(X, tf.float32)

mean_all = tf.reduce_mean(Y)

mean_0 = tf.reduce_mean(Y, axis=0)
mean_1 = tf.reduce_mean(Y, axis=1)
print(mean_all)
print(mean_0)
print(mean_1)

OUT:

参数keepdims设置为True时，保持原来tensor的维度

import tensorflow as tf

X = np.array([[0, 1, 2], [3, 4, 5]])
Y = tf.cast(X, tf.float32)

mean_all = tf.reduce_mean(Y, keepdims=True)

mean_0 = tf.reduce_mean(Y, axis=0, keepdims=True)
mean_1 = tf.reduce_mean(Y, axis=1, keepdims=True)
print(mean_all)
print(mean_0)
print(mean_1)

OUT:

希望这篇文章对大家的学习有所帮助！