To run the analysis, the run_analysis.R file should be executed.
The folder containing the original data should be in the current directory. The file structure of the data directory is below:
UCIHARDataset
test folder
Inertial Signals folder
subject_test
X_test
y_test
train folder
Inertial Signals folder
subject_test
X_test
y_test
activity_labels
features
features_info
README
##Code explaination
The training data are read.
train_data<-read.table(".\\UCIHARDataset\\train\\X_train.txt")
train_label<-read.table(".\\UCIHARDataset\\train\\y_train.txt")
train_subject<-read.table(".\\UCIHARDataset\\train\\subject_train.txt")
train_features<-read.table(".\\UCIHARDataset\\features.txt")
The test data are read.
test_data<-read.table(".\\UCIHARDataset\\test\\X_test.txt")
test_label<-read.table(".\\UCIHARDataset\\test\\y_test.txt")
test_subject<-read.table(".\\UCIHARDataset\\test\\subject_test.txt")
The activity label are read
activity_label<-read.table(".\\UCIHARDataset\\ACTIVITY_LABELS.txt")
Grep is performed to find features where the name ends with either mean() or std()
selectedFeatures<-grepl("^t.*mean()|^t.*std()",train_features$V2,ignore.case=TRUE)
The selected features are stored.
selectedFeaturesList<-train_features[selectedFeatures,]
colTitles<-read.csv('selectedFeatures.csv')
The subject, activity label and the selected data are merged together for training.
train<-cbind(train_subject,train_label,train_data[,colTitles[,1]])
The subject, activity label and the selected data are merged together for training.
test<-cbind(test_subject,test_label,test_data[,colTitles[,1]])
The training and testing data are merged together
data<-rbind(train,test)
The column names are added for the new columns.
titles<-c('subject','activity.label',as.character(colTitles[,3]))
colnames(data)<-titles
The activity label is converted into a factor.
data[,2]<-factor(data[,2],level=activity_label$V1,label=activity_label$V2)
The mean is generated based in grouping of subjects and activities.
tidy_data<-aggregate(data[,3:ncol(data)],by=list(group.subject=data$subject,group.activity=data$activity.label),FUN=mean,na.rm=TRUE)
The data is written to file
write.csv(tidy_data,"tidy_data.txt")
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent