Good Read - Design Patterns

Happy Learning!!!

Day #26 - R - Moving Weighted Average

Example code based on two day workshop on Azure ML module. Simple example storing and accessing data from Azure workspace

	computeAverage<-function(Inputdate)
	{
	print(Inputdate)
	setwd("E:/RNotes/RData/")
	TrafficData <- read.csv(file = "Test3.txt")
	head(TrafficData)
	TrafficData$BusinessDate1 = as.Date(TrafficData$BusinessDate, format = "%Y/%m/%d")
	x = as.Date(Inputdate)-7
	a = TrafficData[TrafficData$BusinessDate1==x,]$TrafficCount
	y = as.Date(x)-14
	b = TrafficData[TrafficData$BusinessDate1==y,]$TrafficCount
	z = as.Date(x)-21
	c = TrafficData[TrafficData$BusinessDate1==z,]$TrafficCount
	wt <- c(.25,.50,.25)
	x <- c(c,b,a)
	print('mean')
	print(mean(x))
	xm <- weighted.mean(x, wt)
	xm
	print('weighted mean')
	print(xm)
	return(xm)
	}
	m=computeAverage('2016-01-29')
	print(m)
	#Save Dataset from below configuration
	#SiteName,BusinessDate,DayofWeek,HourofWeek,TrafficCount
	#Site20,2016/01/01,Monday,14,59
	#Site20,2016/01/08,Monday,14,19
	#Site20,2016/01/15,Monday,14,84
	#Site20,2016/01/22,Monday,14,31

view raw WeightedAverage.R hosted with ❤ by GitHub

	library("AzureML")
	computeAverage<-function(Inputdate,TrafficData)
	{
	head(TrafficData)
	TrafficData$BusinessDate1 = as.Date(TrafficData$BusinessDate, format = "%Y/%m/%d")
	x = as.Date(Inputdate)-7
	a = TrafficData[TrafficData$BusinessDate1==x,]$TrafficCount
	y = as.Date(x)-14
	b = TrafficData[TrafficData$BusinessDate1==y,]$TrafficCount
	z = as.Date(x)-21
	c = TrafficData[TrafficData$BusinessDate1==z,]$TrafficCount
	wt <- c(.25,.50,.25)
	x <- c(c,b,a)
	print('mean')
	print(mean(x))
	xm <- weighted.mean(x, wt)
	xm
	print('weighted mean')
	print(xm)
	return(xm)
	}
	ws <- workspace(
	id = "c54ad6b088114cd5aef68ebc8dedb1d0",
	auth = "eede31e0b9f5463a8c3fcd59a6156a1f",
	api_endpoint = "https://studioapi.azureml.net"
	)
	TrafficData <- download.datasets(
	dataset = ws,
	name = "TrafficDataset"
	)
	m=computeAverage('2016-01-29',TrafficData)
	print(m)

view raw AzureMLExample.R hosted with ❤ by GitHub

Happy Learning!!!

Day #25 - Data Transformations in R

This post is on performing Data Transformations in R. This would be part of feature modelling. Advanced PCA will be done during later stages

	#Data for column attributes
	Y0 = 1:20 + rnorm(100,sd=3)
	Y1 = 1:20 + rnorm(100,sd=2)
	Y2 = 1:20 + rnorm(100,sd=2)
	Y3 = 1:20 + rnorm(100,sd=2)
	Y4 = 1:20 + rnorm(100,sd=2)
	Y5 = 1:20 + rnorm(100,sd=3)
	Y6 = 1:10 + rnorm(100,sd=3)
	Y7 = 1:10 + rnorm(100,sd=2)
	Y8 = 1:20 + rnorm(100,sd=1)
	#Create Data Frame
	data <- data.frame(Y0,Y1,Y2,Y3,Y4,Y5,Y6,Y7,Y8)
	#sample few records
	head(data)
	#Ways to normalize Data
	data$Y0 = (x-min(data$Y0))/(max(data$Y0)-min(data$Y0))
	#Raises Y to the power of 3
	data$Y1 <- (data$Y1)^3
	#Takes the ninth root of Y
	data$Y2 <- (data$Y2)^(1/9)
	#Takes the natural logarithm (ln) of Y
	data$Y3 <- log(data$Y3)
	#Takes the base-10 logarithm of Y
	data$Y4 <- log10(data$Y4)
	#Raises the constant e to the power of Y
	data$Y5 <- exp(data$Y5)
	#Finds the absolute value of Y
	data$Y6 <- abs(data$Y6)
	#Calculates the sine of Y
	data$Y7 <- sin(data$Y7)
	#Calculates the inverse sine (arcsine) of Y
	data$Y8 <- asin(data$Y8)
	#check modified data
	head(data)

view raw DataTransformation.R hosted with ❤ by GitHub

Data Normalization in Python

Happy Learning!!!