mmuratarat

##GRADIENT DESCENT ALGORITHM FOR ONE VARIABLE REGRESSION ANALYSIS

#Data is from an online course taught by Andrew Ng of Stanford University 
#provided by Coursea
#https://github.com/ahawker/machine-learning-coursera

############## Cevrimici (online) bir kaynaktan veri seti okutma ##############
URL_subs <-"https://goo.gl/zEVbcU"
data <- read.table(URL_subs, header=FALSE, sep=",")

mmuratarat

#Fonksiyonu yaratalim
WilcoxonSignedRankTest <- function(x,y){
  diff <- c(x - y)  #her gozlem ikilisi arasindaki farki tutan bir vektor hesaplayalim
  diff.rank <- rank(abs(diff)) #farklarin mutlak degerini alarak siralayalim ve siralarini kaydedelim.
  diff.rank.sign <- diff.rank * sign(diff) #her sira numarasini, karsilik gelen isaret ile etiketleyelim.
  ranks.pos <- sum(diff.rank.sign[diff.rank.sign > 0]) #pozitif degerli siralamalarin sira degerlerini toplayalim.
  ranks.neg <- -sum(diff.rank.sign[diff.rank.sign < 0]) #negatif degerli siralamalarin sira degerlerini toplayalim.
  result<- c(ranks.pos,ranks.neg)
  print(result)
  }

mmuratarat

housing <- read.csv("https://goo.gl/eYbv9F",
                      header = TRUE, 
                      sep = "," )

str(housing)
#'data.frame':	47 obs. of  3 variables:
# $ area    : int  2104 1600 2400 1416 3000 1985 1534 1427 1380 1494 ...
# $ bedrooms: int  3 3 3 2 4 4 3 3 3 3 ...
# $ price   : int  399900 329900 369000 232000 539900 299900 314900 198999 212000 242500 ...

mmuratarat

normalized <- function(x) ( x - mean(x) ) / sd(x)

mmuratarat

GradientDescent <- function(data, alpha, iteration, epsilon){
  
  data <- matrix(unlist(data), ncol=ncol(data), byrow=FALSE)
  
  # bagimli degiskeni ve bagimsiz degiskenleri ayiralim.
  #Veridaki en son kolon, bagimli degiskene ait olmalidir.
  independent.variable<- data[,1:ncol(data)-1]
  dependent.variable<- data[,ncol(data)]
                 
  # girdi degiskenlerine z-değeri normalleştirmesi uygulayalim.

mmuratarat

#Veriyi okutma
housing <- read.csv("https://goo.gl/eYbv9F",
                    header = TRUE, 
                    sep = "," )

#Fonksyonu calistirma
results <- GradientDescent( data = housing, alpha = 0.05, 
                           iteration = 500, epsilon = 0.001)

#Analiz sonuclari

mmuratarat

housing <- read.csv("https://goo.gl/eYbv9F",
                    header = TRUE, 
                    sep = "," )

normalized <- apply(housing[ , -3 ], 2, scale)
normalized_data <- data.frame(cbind(normalized, price = housing$price))
model <- lm( price ~ ., data = normalized_data)
model
#
#Call:

mmuratarat

#Her iterasyon sonrasi elde edilen maliyet fonksiyonu degerlerini cizdirelim.
costs_df <- data.frame( iteration = 1:nrow(results$costs), 
                        costs = results$costs / 1e+8 )

plot(1:nrow(results$costs),costs_df$costs,xlab = "Iteration", ylab = "Cost")

mmuratarat

URL_subs <-"https://goo.gl/zEVbcU"
data <- read.table(URL_subs, header=FALSE, sep=",")

model <- lm(data$V2~ ., data = data)
model
#Call:
#  lm(formula = data$V2 ~ ., data = data)
#
#Coefficients:
#  (Intercept)           V1  

mmuratarat

from __future__ import print_function, division
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt

num_epochs = 100
total_series_length = 50000
truncated_backprop_length = 15
state_size = 4
num_classes = 2