###Set working director as root ###Create file paths ###Load files ###Add columns by combining dataframe columns into single data frame ###Stack test and train data sets - add rows on top of each other to combine test and training data ###Add labels with descriptive variable names by joining on keys with merge ###Only keep columns with "mean" and "std" in their names ###Add descriptive activity names

###Create a second tidy data set as a copy of all.data ###Create a variable containing list grouped by activity and subject ###Apply colmeans function using sapply over all columns and simply result. ###Then transpose to put it in a format easier to validate with after writing out to all.data.step5.txt.

• dir = working directory

• activity.labels.file = points to the file name.txt

• features.file = points to the file name.txt

• test.file= points to the file name.txt

• test.label.file= points to the file name.txt

• test.subject.file= points to the file name.txt

• train.file= points to the file name.txt

• train.label.file= points to the file name.txt

• train.subject.file= points to the file name.txt

• activity.labels.dt = read files in data table

• colnames(activity.labels.dt)<-c("id","activity")

• features.dt= read files into data table

• test.dt= read files into dataframe

• test.label.dt= read files into dataframe

• test.subject.dt= read files into dataframe

• train.dt= read files into dataframe

• train.label.dt= read files into dataframe

• train.subject.dt= read files into dataframe

• test.tmp = temp datatable

• test.all = contains activity label, subject and test data

• train.tmp= temp datatable

• train.all= contains activity label, subject and test data

• all.data = combined test and train data but without proper column names

• features.vector = contains variable names

• features.vector = updated vector with activity and subject column names

• colnames(all.data) = updated all.data column names with columns names

• all.data.subset = only contains mean and standard deviation columns

• all.data.merged = contains descriptive labels for the activities

• all.data2 = contains an independent copy of all.data with test and train data

• data.split = split/grouped by activity id and subject

• all.data.step5<- apply ColMeans function, remove NA's, and transpose the result set (for easy validation in excel) before writing out the result set to a all.data.step5.txt