| title | output |
|---|---|
README |
html_document |
#reads train subject and label data, sets column names train <- read.table("subject_train.txt", header = FALSE, col.names = "subject") trainLabel <- read.table("y_train.txt", header = FALSE, col.names = "activity")
#reads test subject and label data, sets column names test <- read.table("subject_test.txt", header = FALSE, col.names = "subject") testLabel <- read.table("y_test.txt", header = FALSE, col.names = "activity")
#reads variable name file to be used as column names in next step variables <- read.table("features.txt", header = FALSE)
#reads training and test set data, assigns variable names as column names trainX <- read.table("X_train.txt", header = FALSE, col.names = variables[,2]) testX <- read.table("X_test.txt", header = FALSE, col.names = variables[,2])
#column binds the three data frames for both train and test train <- cbind(train, trainLabel, trainX) test <- cbind(test,testLabel, testX)
#row binds the resulting train and test data frames df <- rbind(train, test)
#subsets only the subject, activity, mean(x3) and standard deviation(x3) columns df <- df[,1:8]
#reads activity labels, sets 'activity' column as a factor variable, re-labelling with descriptive labels activityLabels <- read.table("activity_labels.txt") df$activity <- factor(df$activity, levels = activityLabels[,1], labels = activityLabels[,2])
#improves readability of column names names(df) <- gsub("tBodyAcc.mean().", "BodyAccelerationMean",names(df)) names(df) <- gsub("tBodyAcc.std().", "BodyAccelerationStd",names(df))
#creates 2nd independent tidy data frame tidy <- aggregate(df[,3:8], by = list(df$subject,df$activity), FUN = mean, data = df) tidy <- rename(tidy, c("Group.1" ="subject", "Group.2"="activity"))
#writes txt file of tidy data frame write.table(tidy, file="tidy.txt", row.names=FALSE)