README.Rmd

---
title: "README"
author: "Ashok"
date: "November 20, 2015"
output: html_document
---

# Run_Analysis.R Explained

Include the libraries
```{r}
require("dplyr")
```

read the file and list down the files
```{r echo=TRUE}
options(width = 250)
targetfile<-"getdata-projectfiles-UCI HAR Dataset.zip"

#list the files in the zip
unzip(targetfile, list = TRUE)
```

Read the test set and training set files
```{r echo=TRUE}
y_test <- read.table(unz(targetfile, "UCI HAR Dataset/test/y_test.txt"))
x_test <- read.table(unz(targetfile, "UCI HAR Dataset/test/X_test.txt"))
y_train <- read.table(unz(targetfile, "UCI HAR Dataset/train/y_train.txt"))
x_train <- read.table(unz(targetfile, "UCI HAR Dataset/train/X_train.txt"))

subject_test <- read.table(unz(targetfile, "UCI HAR Dataset/test/subject_test.txt"))
subject_train <- read.table(unz(targetfile, "UCI HAR Dataset/train/subject_train.txt"))

features <- read.table(unz(targetfile, "UCI HAR Dataset/features.txt"))
activity_labels <- read.table(unz(targetfile, "UCI HAR Dataset/activity_labels.txt"))
names(activity_labels)<-c('V1', 'Activity')
```


Combine the datasets
```{r echo=TRUE}
x<-rbind(x_train, x_test)
y<-rbind(y_train, y_test)
```

Combine the subject_ids in same order 
```{r echo=TRUE}
subject<-rbind(subject_train, subject_test)
```

Set the names to valid varible names
```{r echo=TRUE}
#names(x)<-make.names(names=features[[2]], unique=TRUE, allow_ = TRUE)
names(x)<-features[[2]]
names(subject)<-"subject_id"
names(activity_labels)<-c('V1','Activity')
```

Add a column with the subject ids merge them
```{r echo=TRUE}
dt<-cbind(subject,x)
```

Merge with activity data with labels
```{r echo=TRUE}
dt<-cbind(y, dt)
dt<-merge(dt, activity_labels, by = 'V1')
```


Now we have all the data, need to filter as per the assignment
```{r echo=TRUE}
reqcols<-names(dt)[grepl(pattern = "std\\(\\)|mean\\(\\)", x = names(dt))]
reqcols <- c(reqcols, "Activity", "subject_id")
dt<- dt[,reqcols]
```

Create tidy data set
```{r echo=TRUE}
#create groups
dt_groups <- group_by(dt, subject_id, Activity)

final <- summarise_each(dt_groups, funs( mean))
```

Write the final dataset to file
```{r echo=TRUE}
write.table(final, file="tidydataset.txt", row.names = FALSE)
```