keras标准化轴参数有什么作用? [英] What does keras normalize axis argument does?

问题描述

我是深度学习的初学者，我正在研究keras中的mnist数据集.

I am a beginner in deep learning and I am working upon the mnist dataset in keras.

我用归一化为

tf.keras.utils.normalize(x_train, axis = 1)

我不明白axis参数的含义.你能帮我这个忙吗?

I don't understand what does the axis argument means. Can you help me out with this?

推荐答案

normalize函数只是执行常规规范化以提高性能:

The normalize function just performs a regular normalization to improve performance:

归一化是对原始范围内的数据进行重新缩放，因此 所有值都在0到1的范围内.

Normalization is a rescaling of the data from the original range so that all values are within the range of 0 and 1.

在另一篇文章中对axis参数有很好的解释:

There is a nice explanation of the axis argument in another post:

axis = -1的含义是什么在keras.argmax中?

例如:

您的数据具有一定的形状(19、19、5、80).这意味着:

Your data has some shape (19,19,5,80). This means:

• 轴= 0-19个元素
• 轴= 1-19个元素
• 轴= 2-5个元素
• 轴= 3-80个元素

此外，对于想深入研究的人，GitHub上的Keras的作者FrançoisChollet有一个解释:

Also, for those who want to go deeper, there is an explanation from François Chollet - Keras’ author- on GitHub:

• 对于密集层，所有RNN层和大多数其他类型的层， 默认应使用axis = -1
• 对于Convolution2D图层 如果使用dim_ordering ="th"(默认值)，则使用axis = 1，
• 对于Convolution2D 具有dim_ordering ="tf"的图层，请使用axis = -1(即默认值).

• For Dense layer, all RNN layers and most other types of layers, the default of axis=-1 is what you should use,
• For Convolution2D layers with dim_ordering="th" (the default), use axis=1,
• For Convolution2D layers with dim_ordering="tf", use axis=-1 (i.e. the default).

https://github.com/fchollet/keras/issues/1921

这篇关于keras标准化轴参数有什么作用?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！