/GettingAndCleaningData_CourseProject

Primary LanguageR

Getting and Cleaning Data Course Project

Prerequisites

  • This program uses data.table for calculating tidy dataset
  • If the data.table is not installed then download the package by executing install.packages("data.table")

Structure of the program

  • The file run_analysis.R program mainly contains following three functions ** Function 'create_complete_model' - generates model that combines all the datasets into a single dataset ** Function 'model_with_mean_and_stddev' - generates model whose column names contain either mean or stddev ** Function 'create_tidy_dataset' - generates model grouping subject_id, activity and taking the mean of all the columns whose column name does not contain mean or stddev ** All the above three functions take logical argument called 'output_model'. If the above methods are called with TRUE then the models are generated as files in the directory named 'output'

How to run the program

  1. Checkout the code to a directory
  2. Open R studio
  3. Set working directory of the R Studio as the directory where the code is checked out.
  4. Source the run_analysis.R file by executing the command source("./run_analysis.R")
  5. For verification the output datasets were checked into the repository for verification.
  6. Before running the program, if the files are already present in the output directory they get overridden.

How to verify the course project

  1. To verify the first question:- to merge all datasets
    • Execute the following command in R studio: model1 = create_complete_model()
    • If we execute the above function with optional argument output_model = TRUE then the output is generated in output directory with file name 'all_observations.csv' str(model1, list.len=999), dim(model1) will give you number of rows, columns, fields of the merged dataset.
  2. To verify the second question:- To extract mean and standard deviation fields
    • Execute the following command in R studio: model2 = model_with_mean_and_stddev()
    • If we execute the above command with output_model = TRUE argument, then the output is generated in the output directory with file name 'observations_with_mean_and_stddev.csv' str(model2), dim(model2) will give number of rows, columns, fields of the dataset that contains only mean and standard deviation fields.
  3. To verify the third question:- use descriptive activity names
    • Both the model1 and model2 variables created in the above commands contain column name 'activity'. Also the output is checked into the github repository in output directory.
    • Output directory contains all_observations.csv with 'activity' column indicating description.
  4. To verify the fifth question:- independent tidy dataset
    • Execute the following command in R studio: model3 = create_tidy_dataset()
    • If the above command is executed with argument output_model = TRUE then the file "tidy_dataset.txt" shall be generated in the output directory.