---
title: "Homework 3: Regression Theory"
output: html_notebook
author: JPeruggia
date: '03/25/2018'
---

** Probelms

**** 1. Data Preprocessing

```{r}

training_data <- read.csv('mnist_train.csv', header=FALSE)
test_data <- read.csv('mnist_test.csv', header=FALSE)
training_df <- as.data.frame(t(training_data))
test_df <- as.data.frame(t(test_data))

# set the name of the column for classifcation value.
names(training_df)[785] <- "Classification"
names(test_df)[785] <- "Classification"

# partition the data.
train_0_1 <- training_df[(training_df$Classification == 0) | (training_df$Classification == 1),]
train_3_5 <- training_df[(training_df$Classification == 3) | (training_df$Classification == 5), ]
test_0_1 <- test_df[(test_df$Classification == 0) | (test_df$Classification == 1),]
test_3_5 <- test_df[(test_df$Classification == 3) | (test_df$Classification == 5), ]

#Dimensions of the Training set with value of 0 or 1
print (dim(train_0_1))
#Dimensions of the Training set with values of 3 or 5
print (dim(train_3_5))
#Dimensions of the Testing set with the values of 0 or 1
print (dim(test_0_1))
#Dimensions of the Testing set with the values of 3 or 5
print (dim(test_3_5))

# seperate the true class lables out.
train_labels_0_1 <- train_0_1$Classification
train_labels_3_5 <- traintrain_3_5_35_df$Classification
test_labels_0_1 <- test_0_1$Classification
test_labels_3_5 <- test_3_5$Classification
# remove them from the original dfs.
train_data_0_1 <- subset(train_0_1, select = names(train_0_1) != "Classification")
train_data_3_5 <- subset(train_3_5, select = names(train_3_5) != "Classification")
test_data_0_1 <- subset(test_0_1, select = names(test_0_1) != "Classification")
test_data_3_5 <- subset(test_3_5, select = names(test_3_5) != "Classification")

# Visualize one image from each class to make sure you have read it in correctly.


m <- matrix(
  unlist(train_data_0_1[!is.na(match(train_labels_0_1, 0)), ][1,]),
  byrow = TRUE,
  nrow = 28,
  ncol = 28
)
t_image = image(z = m, col=gray(0:255/255))
print(t_image)


#image_for_digit(train_01_df,train_01_labels, 0, "Classification of: 0")
#image_for_digit(train_01_df, train_01_labels, 1, "Classification of: 1")
#image_for_digit(train_35_df, train_35_labels, 3, "Classification of: 3")
#image_for_digit(train_35_df, train_35_labels, 5, "Classification of: 5")


```

Add a new chunk by clicking the *Insert Chunk* button on the toolbar or by pressing *Cmd+Option+I*.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the *Preview* button or press *Cmd+Shift+K* to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike *Knit*, *Preview* does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.