Azure ML中对数正态转换的实际计算 [英] Actual calculation for Lognormal transformation in Azure ML

问题描述

我试图了解在对数正态数据转换过程中以0为常数对数据进行归一化的计算方法.

我看到了公式，即z = lognormal.CDF(x ;; mean，std.dev).但是，当我将其应用于Excel电子表格中的数据集时，却没有得到与Azure ML输出匹配的结果.

数据集=( 1.059381 ，0.569366,0.354232,0.012538,0.074269,1.785876,0.611601,0.832213,1.981133,2.428757,0.198116,6.388798， 0.909218,0.125718796,0.348734,0.389039,18.44867,1.270292,3.059661,7.382605,1.422344,0.436206,0.632186,2.214821,0.79113,1.096905,0.415275,0.758696,0.070598,1.052937,0.308737， 2.778732,2.472008,1.192793,1.103876,1.996774,1.33512,1.637168,0.956531,2.893275,0.647775,0.243223,1.563686,0.613898,0.153827,0.485533)

例如我正在考虑数据集中的第一个值

            x实际值= 1.059381475

            x Azure ML处的归一化值= 0.583455024286966

(使用Normlize模型-> LogNormal-> 0作为常量)

            Excel电子表格中的x归一化值= 0.17219

(Lognormal(x，mean，Std.dev，CDF))

感谢您的帮助.

谢谢

Vinodh

解决方案

以下是有关模块工作方式的文档:https://docs.microsoft.com/zh-CN/azure/machine-learning/studio-module-reference/normalize-data

此致，
Jaya

Hi ,

I am trying to understand the calculation used to normalize the data during the 'Lognormal' data transformation process with 0 as constant.

I see the formula i.e. z = lognormal.CDF(x;;mean,std.dev). However when i apply the same to the dataset in a excel spreadsheet i am not getting the results that matches Azure ML output.

Dataset = (1.059381,0.569366,0.354232,0.012538,0.074269,1.785876,0.611601,0.832213,1.981133,2.428757,0.198116,6.388798,0.909218,0.125718796,0.348734,0.389039,18.44867,1.270292,3.059661,7.382605,1.422344,0.436206,0.632186,2.214821,0.79113,1.096905,0.415275,0.758696,0.070598,1.052937,0.308737, 2.778732,2.472008,1.192793,1.103876,1.996774,1.33512,1.637168,0.956531,2.893275,0.647775,0.243223,1.563686,0.613898,0.153827,0.485533)

for e.g. i am considering 1st value in data set

            x actual value = 1.059381475

            x normalized value at Azure ML = 0.583455024286966

(using Normlize model - >LogNormal->0 as constant) 

            x normalized value at Excel spreadsheet = 0.17219

(Lognormal(x,mean, Std.dev, CDF))

Appreciate your help.

Thanks,

Vinodh

解决方案

Hi, 

Here is the documentation on how the module works: https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/normalize-data

Regards,
Jaya

这篇关于Azure ML中对数正态转换的实际计算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！