Create run_analysis.R

This script is supposed to work, conditioning you have UCI HAR Dataset extracted to the same directory. The source of this dataset is here:
https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip

run_analysis.R

@@ -0,0 +1,37 @@
+# R script for Getting and Cleaning Data Project
+
+# 1. Preparing column for "Subject" variable
+subj_train <- read.table("UCI HAR Dataset/train/subject_train.txt", header=F)
+subj_test <- read.table("UCI HAR Dataset/test/subject_test.txt", header=F)
+subj <- rbind(subj_train, subj_test)
+names(subj) <- "subject"
+
+# 2. Preparing columns for "features" variables
+X_train <- read.table("UCI HAR Dataset/train/X_train.txt", header=F)
+X_test <- read.table("UCI HAR Dataset/test/X_test.txt", header=F)
+X <- rbind(X_train, X_test)
+feat <- read.table("UCI HAR Dataset/features.txt", header=F)
+feat_ind <- grep("-mean\\(\\)|-std\\(\\)", feat[, 2])
+X <- X[, feat_ind]
+names(X) <- feat[feat_ind, 2]
+names(X) <- gsub("\\(|\\)", "", names(X))
+
+# 3. Preparing column for "activity" variable
+y_train <- read.table("UCI HAR Dataset/train/y_train.txt", header=F)
+y_test <- read.table("UCI HAR Dataset/test/y_test.txt", header=F)
+Y <- rbind(y_train, y_test)
+activ <- read.table("UCI HAR Dataset/activity_labels.txt", header=F)
+activ[, 2] = gsub("_", "", as.character(activ[, 2]))
+Y[,1] = activ[Y[,1], 2]
+names(Y) <- "activity"
+
+# 4. Combining all the variables into one dataset
+dataset <- cbind(subj, Y, X)
+write.csv(dataset, "dataset.csv")
+
+# 5. Creating a 2nd tidy data set with the average of each variable for each activity and each subject.
+library(reshape)
+ids = c("SubjID", "ActID", "ActName")
+melted <- melt(dataset, id.vars = c("subject", "activity"))
+avg_dataset <- cast(subject + variable ~ activity, data = melted, fun=mean)
+write.csv(avg_dataset, "avg_data.csv")