Book a Demo!
CoCalc Logo Icon
StoreFeaturesDocsShareSupportNewsAboutPoliciesSign UpSign In
debakarr
GitHub Repository: debakarr/machinelearning
 Path: blob/master/Part 9 - Dimension Reduction/Linear Discriminant Analysis/lda.R
1339 views
1
# LDA

2


3
# Importing the dataset

4
dataset = read.csv('Wine.csv')

5


6
# Splitting the dataset into the Training set and Test set

7
# install.packages('caTools')

8
library(caTools)

9
set.seed(123)

10
split = sample.split(dataset$Customer_Segment, SplitRatio = 0.8)

11
training_set = subset(dataset, split == TRUE)

12
test_set = subset(dataset, split == FALSE)

13


14
# Feature Scaling

15
training_set[-14] = scale(training_set[-14])

16
test_set[-14] = scale(test_set[-14])

17


18
# Applying LDA

19
library(MASS)

20
lda = lda(formula = Customer_Segment ~ ., data = training_set)

21
training_set = as.data.frame(predict(lda, training_set))

22
training_set = training_set[c(5, 6, 1)]

23
test_set = as.data.frame(predict(lda, test_set))

24
test_set = test_set[c(5, 6, 1)]

25


26
# Fitting SVM to the Training set

27
# install.packages('e1071')

28
library(e1071)

29
classifier = svm(formula = class ~ .,

30
                 data = training_set,

31
                 type = 'C-classification',

32
                 kernel = 'linear')

33


34
# Predicting the Test set results

35
y_pred = predict(classifier, newdata = test_set[-3])

36


37
# Making the Confusion Matrix

38
cm = table(test_set[, 3], y_pred)

39


40
# Visualising the Training set results

41
library(ElemStatLearn)

42
set = training_set

43
X1 = seq(min(set[, 1]) - 1, max(set[, 1]) + 1, by = 0.01)

44
X2 = seq(min(set[, 2]) - 1, max(set[, 2]) + 1, by = 0.01)

45
grid_set = expand.grid(X1, X2)

46
colnames(grid_set) = c('x.LD1', 'x.LD2')

47
y_grid = predict(classifier, newdata = grid_set)

48
plot(set[, -3],

49
     main = 'SVM (Training set)',

50
     xlab = 'LD1', ylab = 'LD2',

51
     xlim = range(X1), ylim = range(X2))

52
contour(X1, X2, matrix(as.numeric(y_grid), length(X1), length(X2)), add = TRUE)

53
points(grid_set, pch = '.', col = ifelse(y_grid == 2, 'deepskyblue', ifelse(y_grid == 1, 'springgreen3', 'tomato')))

54
points(set, pch = 21, bg = ifelse(set[, 3] == 2, 'blue3', ifelse(set[, 3] == 1, 'green4', 'red3')))

55


56
# Visualising the Test set results

57
library(ElemStatLearn)

58
set = test_set

59
X1 = seq(min(set[, 1]) - 1, max(set[, 1]) + 1, by = 0.01)

60
X2 = seq(min(set[, 2]) - 1, max(set[, 2]) + 1, by = 0.01)

61
grid_set = expand.grid(X1, X2)

62
colnames(grid_set) = c('x.LD1', 'x.LD2')

63
y_grid = predict(classifier, newdata = grid_set)

64
plot(set[, -3], main = 'SVM (Test set)',

65
     xlab = 'LD1', ylab = 'LD2',

66
     xlim = range(X1), ylim = range(X2))

67
contour(X1, X2, matrix(as.numeric(y_grid), length(X1), length(X2)), add = TRUE)

68
points(grid_set, pch = '.', col = ifelse(y_grid == 2, 'deepskyblue', ifelse(y_grid == 1, 'springgreen3', 'tomato')))

69
points(set, pch = 21, bg = ifelse(set[, 3] == 2, 'blue3', ifelse(set[, 3] == 1, 'green4', 'red3')))

70