Homework.Rmd

---
title: "Homework"
author: "Chenye Xu"
date: "2018???9???16???"
output: html_document
---

```
knitr::opts_chunk$set(echo = TRUE)
```

## how you built your model

Seprate into 5 classes as A B C D E Observe the different outcomes chooseing different variables.

```
trainRaw <- read.csv("pml-training.csv")
testRaw <- read.csv("pml-testing.csv")

dim(trainRaw)
dim(testRaw)
sum(complete.cases(trainRaw))

trainRaw <- trainRaw[, colSums(is.na(trainRaw)) == 0] 
testRaw <- testRaw[, colSums(is.na(testRaw)) == 0]
classe <- trainRaw$classe
trainRemove <- grepl("^X|timestamp|window", names(trainRaw))
trainRaw <- trainRaw[, !trainRemove]
trainCleaned <- trainRaw[, sapply(trainRaw, is.numeric)]
trainCleaned$classe <- classe
testRemove <- grepl("^X|timestamp|window", names(testRaw))
testRaw <- testRaw[, !testRemove]
testCleaned <- testRaw[, sapply(testRaw, is.numeric)]
set.seed(123456) 
inTrain <- createDataPartition(trainCleaned$classe, p=0.75, list=F)
trainData <- trainCleaned[inTrain, ]
testData <- trainCleaned[-inTrain, ]
controlRf <- trainControl(method="cv", 5)
modelRf <- train(classe ~ ., data=trainData, method="rf", trControl=controlRf, ntree=250)
modelRf
```

## how you used cross validation

Clean the raw data and split it into two set for training(75%) and testing(25%).

```
inTrain <- createDataPartition(trainCleaned$classe, p=0.75, list=F)
trainData <- trainCleaned[inTrain, ]
testData <- trainCleaned[-inTrain, ]
```

## what you think the expected out of sample error is

Corresponding to the quantity: 1-accuracy in the cross-validation data.



## why you made the choices you did

There are many NA in data, and the classe is a variable with no order.

```
corrPlot <- cor(trainData[, -length(names(trainData))])
corrplot(corrPlot, method="color")
treeModel <- rpart(classe ~ ., data=trainData, method="class")
prp(treeModel)
```