In mertayD/Coding_Project_1_NN: coding project nn

itle: "Report for Project One: Nearest Neighbors" author: "Kaitlyn Grubb" date: "r Sys.Date()" output: rmarkdown::html_vignette vignette: > %\VignetteIndexEntry{Report for Project One: Nearest Neighbors} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8}

---

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

data.name.vec <- c(
  "spam",
  "SAheart",
  "zip.train",
  "prostate",
  "ozone")
data.list <- list()
library(ElemStatLearn)
for(data.name in data.name.vec)
{
  data(list = data.name, package="ElemStatLearn")
  data.list[[data.name]] <- get(data.name)
}
str(data.list)

is.binary <- ElemStatLearn::zip.train[,1] %in% c(0,1)

data.list <- list(
  spam=list(
    label.vec=ifelse(ElemStatLearn::spam$spam=="spam", 1, 0),
    feature.mat=as.matrix(ElemStatLearn::spam[, -ncol(ElemStatLearn::spam)])),
  zip.train=list(
    label.vec=ElemStatLearn::zip.train[is.binary,1],
    feature.mat=ElemStatLearn::zip.train[is.binary,-1])
)


n.folds <- 5
resultsList <- list()
for(data.name in names(data.list))
{
  one.data.set <- data.list[[data.name]]
  set.seed(1)
  fold.vec <- sample(rep(1:n.folds, l=nrow(one.data.set$feature.mat)))
    #NearestNeighborsCV(one.data.set$feature.mat,   one.data.set$label.vec)
  for(test.fold in 1:(n.folds))
  {
    is.train <- fold.vec != test.fold
    fit <- NNLearnCV(one.data.set$feature.mat[is.train],   one.data.set$label.vec[is.train])
    pred.vec <- fit$predict(one.data.set$feature.mat[!is.train])
    ## compute test error of pred vec with respect to one data.set$label[!is.train]

    # LIST OF LISTS
    resultsList[data.name] <- NNLearnCV(data.name[ , -1], data.name[ , 1], max.neighbors, fold.vec, n.folds)# save your results
  }
}

Data set 1: spam

Matrix of loss values

#print out and/or plot the matrix.
pander::pandoc.table(resultsList[1]$train.loss.mat)

comment on difference between NN and baseline.

Train/validation loss plot

#plot the two loss functions.
ggplot(resultsList[1]$train.loss.vec)
ggplot(resultsList[1]$validation.loss.vec)

What is the optimal number of neighbors?

#plot the two loss functions.
resultsList[1]$selected.neighbors

Data set 2: SAheart

Matrix of loss values

#print out and/or plot the matrix.
pander::pandoc.table(resultsList[2]$train.loss.mat)

comment on difference between NN and baseline.

Train/validation loss plot

#plot the two loss functions.
ggplot(resultsList[2]$train.loss.vec)
ggplot(resultsList[2]$validation.loss.vec)

What is the optimal number of neighbors?

#plot the two loss functions.
resultsList[2]$selected.neighbors

Data set 3: zip.train

Matrix of loss values

#print out and/or plot the matrix.
pander::pandoc.table(resultsList[3]$train.loss.mat)

comment on difference between NN and baseline.

Train/validation loss plot

#plot the two loss functions.
ggplot(resultsList[3]$train.loss.vec)
ggplot(resultsList[3]$validation.loss.vec)

What is the optimal number of neighbors?

#plot the two loss functions.
resultsList[3]$selected.neighbors

Data set 4: prostate

Matrix of loss values

#print out and/or plot the matrix.
pander::pandoc.table(resultsList[4]$train.loss.mat)

comment on difference between NN and baseline.

Train/validation loss plot

#plot the two loss functions.
ggplot(resultsList[4]$train.loss.vec)
ggplot(resultsList[4]$validation.loss.vec)

What is the optimal number of neighbors?

#plot the two loss functions.
resultsList[4]$selected.neighbors

Data set 5: ozone

Matrix of loss values

#print out and/or plot the matrix.
pander::pandoc.table(resultsList[5]$train.loss.mat)

comment on difference between NN and baseline.

Train/validation loss plot

#plot the two loss functions.
ggplot(resultsList[5]$train.loss.vec)
ggplot(resultsList[5]$validation.loss.vec)

What is the optimal number of neighbors?

#plot the two loss functions.
resultsList[5]$selected.neighbors

mertayD/Coding_Project_1_NN documentation built on June 1, 2019, 3:57 a.m.