coppeliaMLA

#First we are going to set up probaility distributions for our beliefs about the inputs
#We've been told ARPU is about £7 and it's very unlikely to be higher than £10 or lower than £4
#So we'll go for a normal distribution centred at 7 with 5% and 95% quantiles at 4 and 10

#Show how we get the variance

arpu.sd<-3/1.96
x<-seq(0, 15,by=0.5)
d<-dnorm(x, 7, arpu.sd)
plot(x, d, type='l')

coppeliaMLA

#Load libraries

library(rjson)
library(stringr)


dfToJSON<-function(df, mode='vector'){
  
  colToList<-function(x, y){
    

coppeliaMLA

'''
Created on Feb 6, 2014

@author: sraper
'''

import itertools, urllib, urllib2, time, re, random
from bs4 import BeautifulSoup

def catchURL(queryURL): # Nicked this from someone. Afraid I can't remember who. Sorry

coppeliaMLA

import os

dir = 'put your director in here'

for filename in os.listdir(dir):
    with open(dir+filename, 'r') as fin:
        data = fin.read().splitlines(True)
    with open(dir+filename, 'w') as fout:
        fout.writelines(data[1:])

coppeliaMLA

    drop table if exists recommender_set_num; --In case you need to rerun the script
    drop table if exists person_ids_full_names; 
    drop table if exists recom_names;
    
    -- Set up a table to load the recommendations data into
    create external table if not exists recommender_set_num
    (
     userID     bigint,
     itemID     bigint
    ) row format delimited fields terminated by ','

coppeliaMLA

x<-seq(1,10, by=0.1)
y<-sin(x/4)+rnorm(91, 0,0.05) #Sin fucntion plus noise
plot(x,y)


#Knots at 2, 4 and 6
x2<-x^2  
x3<-x^3
k1<-(x>2)*(x-2)^3
k2<-(x>4)*(x-4)^3

coppeliaMLA

import os, csv

progDir = '/pathToFolderContainingCSVs/'



for filename in os.listdir(progDir):
    if filename != '.DS_Store':
        with open(progDir+filename, 'rb') as csvfile:
            progReader = csv.reader(csvfile, delimiter=',', quotechar='"')

coppeliaMLA

modBin<-function(k, n, p){
  
  if (k<=n) {
    return(dbinom(k, n, p))
  }
  else {
    return(0)
  }
}

coppeliaMLA

#Sequence for adding new data
s<-seq(20,50, by=5)

#Set up object for recording clusters
clus.change<-NULL

#Cycle through the clustering solutions
for (i in s){
 
  hc <- hclust(dist(USArrests[1:i,]), "ave")

coppeliaMLA

#Run hclust
hc <- hclust(dist(USArrests[1:40,]), "ave")

#Function for extracting nodes and links
extractGraph<-function(hc){
  
  n<-length(hc$order)
  m<-hc$merge

  links<-data.frame(source=as.numeric(), target=as.numeric(), value=as.numeric())