dgrapov

#SOM example using wines data set
library(kohonen)
data(wines)
set.seed(7)

#create SOM grid
sommap <- som(scale(wines), grid = somgrid(2, 2, "hexagonal"))

## use hierarchical clustering to cluster the codebook vectors
groups<-3

dgrapov

#initialize
library(shiny)
library(ggplot2)
library(purrr)
library(dplyr)


#example data
data(iris)

dgrapov

#plotly box or lasso select linked to 
# DT data table
# using Wage data
# the out group: is sex:Male, region:Middle Atlantic +


library(ggplot2)
library(plotly)
library(dplyr) 
library(ISLR) 

dgrapov

DATA SCIENCE EXERCISE

The following challenge requires the beer reviews data set called beer_reviews.csv. This data set can be downloaded from the following site: https://data.world/socialmediadata/beeradvocate . Note you can create a free temporary account to download this .csv.

Questions to answer using this data:
Which brewery produces the strongest beers by ABV%?
If you had to pick 3 beers to recommend using only this data, which would you pick?
Which of the factors (aroma, taste, appearance, palette) are most important in determining the overall quality of a beer?

Additional math/coding question unrelated to the data:

dgrapov

#see updated code base and some examples in the function "test" 
# https://github.com/dgrapov/devium/blob/master/R/Devium%20PLS%20%20and%20OPLS.r

#Orthogonal Signal Correction for PLS models (OPLS)
#adapted from an example in the book <a href="http://www.springer.com/life+sciences/systems+biology+an+bioinfomatics/book/978-3-642-17840-5">"Chemometrics with R by Ron Wehrens"</a>

#this code requires the following packages:
need.packages<-c("pls", # to generate PLS models
"ggplot2" ) # to plot results

dgrapov

#check for and/or install dependencies
need<-c("RCurl","ggplot2","gridExtra","reshape2")
for(i in 1:length(need)){
  if(require(need[i], character.only = TRUE)==FALSE){ install.packages(need[i]);library(need[i], character.only = TRUE)} else { library(need[i],character.only = TRUE)}
}

if(require(pcaMethods)==FALSE){
  need<-c('Rcpp', 'rJava',
  'Matrix', 'cluster', 'foreign', 'lattice', 'mgcv', 'survival')
for(i in 1:length(need)){

dgrapov

> in
Error: unexpected 'in' in "in"

dgrapov

# Basic PCA example
# use www.createdatasol.com for 
# an advanced user interface 

#required packages for plotting
library(ggplot2)
library(ggrepel)

#load data
data<-read.csv('~/Sampledata.csv', 

dgrapov

#needed packages
library(andrews)
library(ggplot2)

#get some data here I use "mtcars"
data<-mtcars
fct<-as.factor(mtcars$cyl) # factor for intial grouping 


#get andrews encoding (x and y coords)

dgrapov

#' @title fast_tanimoto 
#' @param mat matrix or data frame of numeric values
#' @param output 'matrix' (default) or 'edge list' (non-redundant and undirected)
#' @param progress TRUE, show progress
#' @imports reshape2
fast_tanimoto<-function(mat,output='matrix',progress=TRUE){
  mat[is.na(mat)]<-0 
  
  #scoring function
  score<-function(x){sum(x==2)/sum(x>0)}