Styjun

Most elegant way to write a one shot IF

Spicket or spigot?

What is a macro? Difference between macro and function?

Is this hogweed?

In native German words, is Q always followed by U, as in English?

Prime parity peregrination

How can I convince my reader that I will not use a certain trope?

What is the line crossing the Pacific Ocean that is shown on maps?

Using aluminium busbar/cables in an aircraft instead of copper

How to formulate maximum function in a constraint?

Generate and graph the Recamán Sequence

How did researchers use to find articles before the Internet and the computer era?

Averting Real Women Don’t Wear Dresses

What is "oversubscription" in Networking?

Can another character physically take something that Mage Hand is carrying/holding?

Who gets an Apparition licence?

Plotting the gradient descent

How exactly is a normal force exerted, at the molecular level?

Why was Mal so quick to drop Bester in favour of Kaylee?

What exactly is a fey/fiend/celestial spirit?

Why isn’t the tax system continuous rather than bracketed?

"Plugged in" or "Plugged in in"

Can a police officer film me on their personal device in my own home?

Different budgets within roommate group

What columns should use in order to train Random Forest?

R - Random Forest and more than 53 categoriesR Random Forests Variable ImportanceSuggestions for speeding up Random ForestsR random forest - training set using target column for predictionCreating a loop for different random forest training algorithmsTrain a random forest algorithm using various columnsRandom forest bootstrap training and forest generationDeleting rows in training dataset for Random ForestTraining a Random Forest on TensorflowError in Bagging with party::cforestWhy would different random forest implementations in R yield different results?

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;

0

Background

I am new in machine learning. I want to train model by using Random Forest algorithm. I have database which includes total 9 columns, in which 8 are independent variables and last (9th) variable 'Class' is dependent variable. Dependent variable is predictor variable which contains 3 categories i.e. - S, N, R. All independent variables (except 2) contain more categories than 53. Code shows error when categories become more than 53. I want to train the model in order to identify whether the database line is Suspicious (S), Normal (N), Robot (R). Column numbers 4 and 7 contain more than 19k categories/levels. These are the important columns because they contain attack entries/features etc. How to derive other variables from them becomes complicated.

Code

library('ROCR')
library('randomForest')
library('caret')
library('ranger')

database<-read.csv('data1.csv')
set.seed(1000)
train<-sample(1:310341,217239,replace = FALSE)
traindata<-database[train,]
testdata<-database[-train,]
# fit <- train(database$Class ~ ., data = traindata, method = "ranger")
fit<-randomForest(Class~.,data = traindata, ntree=500, importance= TRUE, proximity = TRUE, na.action = na.roughfix)

Efforts taken

I have tried above code, but due to 4th, and 7th column, it shows error of "more than 53 columns can not be used".

Any help would be appreciated to resolve this issue...

r random-forest

share|improve this question

edited Mar 26 at 18:22

DSD

asked Mar 25 at 11:06

DSDDSD

122 bronze badges

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  Possible duplicate of R - Random Forest and more than 53 categories
  
  – divibisan
  Apr 2 at 17:49
  

  
  
  
  

add a comment | 

0

Background

I am new in machine learning. I want to train model by using Random Forest algorithm. I have database which includes total 9 columns, in which 8 are independent variables and last (9th) variable 'Class' is dependent variable. Dependent variable is predictor variable which contains 3 categories i.e. - S, N, R. All independent variables (except 2) contain more categories than 53. Code shows error when categories become more than 53. I want to train the model in order to identify whether the database line is Suspicious (S), Normal (N), Robot (R). Column numbers 4 and 7 contain more than 19k categories/levels. These are the important columns because they contain attack entries/features etc. How to derive other variables from them becomes complicated.

Code

library('ROCR')
library('randomForest')
library('caret')
library('ranger')

database<-read.csv('data1.csv')
set.seed(1000)
train<-sample(1:310341,217239,replace = FALSE)
traindata<-database[train,]
testdata<-database[-train,]
# fit <- train(database$Class ~ ., data = traindata, method = "ranger")
fit<-randomForest(Class~.,data = traindata, ntree=500, importance= TRUE, proximity = TRUE, na.action = na.roughfix)

Efforts taken

I have tried above code, but due to 4th, and 7th column, it shows error of "more than 53 columns can not be used".

Any help would be appreciated to resolve this issue...

r random-forest

share|improve this question

edited Mar 26 at 18:22

DSD

asked Mar 25 at 11:06

DSDDSD

122 bronze badges

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  Possible duplicate of R - Random Forest and more than 53 categories
  
  – divibisan
  Apr 2 at 17:49
  

  
  
  
  

add a comment | 

Background

I am new in machine learning. I want to train model by using Random Forest algorithm. I have database which includes total 9 columns, in which 8 are independent variables and last (9th) variable 'Class' is dependent variable. Dependent variable is predictor variable which contains 3 categories i.e. - S, N, R. All independent variables (except 2) contain more categories than 53. Code shows error when categories become more than 53. I want to train the model in order to identify whether the database line is Suspicious (S), Normal (N), Robot (R). Column numbers 4 and 7 contain more than 19k categories/levels. These are the important columns because they contain attack entries/features etc. How to derive other variables from them becomes complicated.

Code

library('ROCR')
library('randomForest')
library('caret')
library('ranger')

database<-read.csv('data1.csv')
set.seed(1000)
train<-sample(1:310341,217239,replace = FALSE)
traindata<-database[train,]
testdata<-database[-train,]
# fit <- train(database$Class ~ ., data = traindata, method = "ranger")
fit<-randomForest(Class~.,data = traindata, ntree=500, importance= TRUE, proximity = TRUE, na.action = na.roughfix)

Efforts taken

I have tried above code, but due to 4th, and 7th column, it shows error of "more than 53 columns can not be used".

Any help would be appreciated to resolve this issue...

r random-forest

share|improve this question

edited Mar 26 at 18:22

DSD

asked Mar 25 at 11:06

DSDDSD

122 bronze badges

library('ROCR')
library('randomForest')
library('caret')
library('ranger')

database<-read.csv('data1.csv')
set.seed(1000)
train<-sample(1:310341,217239,replace = FALSE)
traindata<-database[train,]
testdata<-database[-train,]
# fit <- train(database$Class ~ ., data = traindata, method = "ranger")
fit<-randomForest(Class~.,data = traindata, ntree=500, importance= TRUE, proximity = TRUE, na.action = na.roughfix)

share|improve this question

edited Mar 26 at 18:22

DSD

asked Mar 25 at 11:06

DSDDSD

122 bronze badges

share|improve this question

edited Mar 26 at 18:22

DSD

asked Mar 25 at 11:06

DSDDSD

122 bronze badges

asked Mar 25 at 11:06

DSDDSD

122 bronze badges

asked Mar 25 at 11:06

DSDDSD

122 bronze badges