API tools faq
paste
Guest User

Untitled

a guest
Nov 18th, 2017
106
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 4.40 KB | None | 0 0
raw download clone embed print report
  1. ###### MICE PACKAGE ######
  2. library(mice)
  3. library(class)
  4. library(missForest)
  5.  
  6. MyData <- iris
  7. #generate random "2%" missing values
  8. MissingData <- prodNA(MyData[-5], noNA = 0.02)
  9. #using mic to impute iris Data set
  10. #using method pmm (Predictive Mean Matching) for numeric variables
  11. ImputedData <- mice(MissingData, m=3, maxit = 20, method = 'pmm', seed = 222)
  12. MytestData <- complete(ImputedData,1) #testdata
  13. #using k-NN classifier
  14. predict_species_class<- knn(train = MyData[-5], test = MytestData, cl=MyData$Species , k = 3,prob=TRUE)
  15. #using confusion matrix to look the performance of k-NN classifier
  16. confusionMatrix(MyData$Species,predict_species_class)
  17. #Root mean square error between imputed Data and MytestData
  18. RMSE(MyData[-5],MytestData)
  19.  
  20.  
  21. ###### MICE PACKAGE ######
  22. library(mice)
  23. library(class)
  24. library(missForest)
  25.  
  26. MyData <- iris
  27. #generate random "5%" missing values
  28. MissingData <- prodNA(MyData[-5], noNA = 0.05)
  29. #using mic to impute iris Data set
  30. #using method pmm (Predictive Mean Matching) for numeric variables
  31. ImputedData <- mice(MissingData, m=3, maxit = 20, method = 'pmm', seed = 222)
  32. MytestData <- complete(ImputedData,1) #testdata
  33. #using k-NN classifier
  34. predict_species_class<- knn(train = MyData[-5], test = MytestData, cl=MyData$Species , k = 3,prob=TRUE)
  35. #using confusion matrix to look the performance of k-NN classifier
  36. confusionMatrix(MyData$Species,predict_species_class)
  37. #Root mean square error between imputed Data and MytestData
  38. RMSE(MyData[-5],MytestData)
  39.  
  40.  
  41. ###### MICE PACKAGE ######
  42. library(mice)
  43. library(class)
  44. library(missForest)
  45.  
  46. MyData <- iris
  47. #generate random "10%" missing values
  48. MissingData <- prodNA(MyData[-5], noNA = 0.1)
  49. #using mic to impute iris Data set
  50. #using method pmm (Predictive Mean Matching) for numeric variables
  51. ImputedData <- mice(MissingData, m=3, maxit = 20, method = 'pmm', seed = 222)
  52. MytestData <- complete(ImputedData,1) #testdata
  53. #using k-NN classifier
  54. predict_species_class<- knn(train = MyData[-5], test = MytestData, cl=MyData$Species , k = 3,prob=TRUE)
  55. #using confusion matrix to look the performance of k-NN classifier
  56. confusionMatrix(MyData$Species,predict_species_class)
  57. #Root mean square error between imputed Data and MytestData
  58. RMSE(MyData[-5],MytestData)
  59.  
  60.  
  61. ###### MICE PACKAGE ######
  62. library(mice)
  63. library(class)
  64. library(missForest)
  65.  
  66. MyData <- iris
  67. #generate random "15%" missing values
  68. MissingData <- prodNA(MyData[-5], noNA = 0.15)
  69. #using mic to impute iris Data set
  70. #using method pmm (Predictive Mean Matching) for numeric variables
  71. ImputedData <- mice(MissingData, m=3, maxit = 20, method = 'pmm', seed = 222)
  72. MytestData <- complete(ImputedData,1) #testdata
  73. #using k-NN classifier
  74. predict_species_class<- knn(train = MyData[-5], test = MytestData, cl=MyData$Species , k = 3,prob=TRUE)
  75. #using confusion matrix to look the performance of k-NN classifier
  76. confusionMatrix(MyData$Species,predict_species_class)
  77. #Root mean square error between imputed Data and MytestData
  78. RMSE(MyData[-5],MytestData)
  79.  
  80.  
  81. ###### MICE PACKAGE ######
  82. library(mice)
  83. library(class)
  84. library(missForest)
  85.  
  86. MyData <- iris
  87. #generate random "20%" missing values
  88. MissingData <- prodNA(MyData[-5], noNA = 0.2)
  89. #using mic to impute iris Data set
  90. #using method pmm (Predictive Mean Matching) for numeric variables
  91. ImputedData <- mice(MissingData, m=3, maxit = 20, method = 'pmm', seed = 222)
  92. MytestData <- complete(ImputedData,1) #testdata
  93. #using k-NN classifier
  94. predict_species_class<- knn(train = MyData[-5], test = MytestData, cl=MyData$Species , k = 3,prob=TRUE)
  95. #using confusion matrix to look the performance of k-NN classifier
  96. confusionMatrix(MyData$Species,predict_species_class)
  97. #Root mean square error between imputed Data and MytestData
  98. RMSE(MyData[-5],MytestData)
  99.  
  100.  
  101.  
  102.  
  103. ###### MICE PACKAGE ######
  104. library(mice)
  105. library(class)
  106. library(missForest)
  107.  
  108. MyData <- iris
  109. #generate random "25%" missing values
  110. MissingData <- prodNA(MyData[-5], noNA = 0.25)
  111. #using mic to impute iris Data set
  112. #using method pmm (Predictive Mean Matching) for numeric variables
  113. ImputedData <- mice(MissingData, m=3, maxit = 20, method = 'pmm', seed = 222)
  114. MytestData <- complete(ImputedData,1) #testdata
  115. #using k-NN classifier
  116. predict_species_class<- knn(train = MyData[-5], test = MytestData, cl=MyData$Species , k = 3,prob=TRUE)
  117. #using confusion matrix to look the performance of k-NN classifier
  118. confusionMatrix(MyData$Species,predict_species_class)
  119. #Root mean square error between imputed Data and MytestData
  120. RMSE(MyData[-5],MytestData)
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!