Re: [R] RandomForest tuning the parameters

2023-05-13 Thread varin sacha via R-help
Dear Eric thanks for your code, Dear R-experts, Here below my R code, still the same error message at the end while trying to tune the parameters!  Any hints to make it work would be appreciated. Best, # Create features a

Re: [R] RandomForest tuning the parameters

2023-05-09 Thread Eric Berger
Hi Sacha, On second thought, perhaps this is more the direction that you want ... X2 = cbind(X_train,y_train) colnames(X2)[3] = "y" regr2<-randomForest(y~x1+x2, data=X2,maxnodes=10, ntree=10) regr regr2 #Make prediction predictions= predict(regr, X_test) predictions2= predict(regr2, X_test) HTH,

Re: [R] RandomForest tuning the parameters

2023-05-08 Thread Eric Berger
Hi, One problem you have is with the command: regr<-randomForest(y~x1+x2, data=X_train, proximity=TRUE) What you need is something like this: X2 <- cbind(X,y) regr<-randomForest(y~x1+x2, data=X2, proximity=TRUE) HTH, Eric On Mon, May 8, 2023 at 11:11 PM varin sacha via R-help wrote: > > Dear

[R] RandomForest tuning the parameters

2023-05-08 Thread varin sacha via R-help
Dear R-experts, Here below a toy example with some error messages, especially at the end of the code (Tuning the parameters). Your help to correct my R code would be highly appreciated. ### #libraries library(lattice) library(ggplot2) library(caret) library(

Re: [R] randomForest out of bag prediction

2019-01-13 Thread Michael Mayer
accuracies are not unexpected. Gesendet von Mail für Windows 10 Von: Witold E Wolski Gesendet: Samstag, 12. Januar 2019 18:56 An: r-help@r-project.org Betreff: [R] randomForest out of bag prediction Hello, I am just not sure what the predict.RandomForest function is doing... I confused. I would expect

Re: [R] randomForest out of bag prediction

2019-01-12 Thread Peter Langfelder
See inline. On Sat, Jan 12, 2019 at 9:56 AM Witold E Wolski wrote: > ypred_oob <- predict(diachp.rf) AFAIK these are, indeed, the out-of-bag predictions. > dataX <- data %>% select(-quality) # remove response. > ypred <- predict( diachp.rf, dataX ) These are not out of bag predictions. dataX

Re: [R] randomForest out of bag prediction

2019-01-12 Thread Bert Gunter
Off topic. But see here: https://stats.stackexchange.com/questions/61405/random-forest-and-prediction -- Bert Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Sa

[R] randomForest out of bag prediction

2019-01-12 Thread Witold E Wolski
Hello, I am just not sure what the predict.RandomForest function is doing... I confused. I would expect the predictions for these 2 function calls to predict the same: ```{r} diachp.rf <- randomForest(quality~.,data=data,ntree=50, importance=TRUE) ypred_oob <- predict(diachp.rf) dataX <- data %>

[R] RandomForest-question about split point and GetTree function

2016-05-06 Thread Adam Sanders
I am looking at results of a random forest. In the documentation, it says the following for categorical variables: "For categorical predictors, the splitting point is represented by an integer, whose binary expansion gives the identities of the categories that goes to left or right. For example, i

[R] randomForest and factor predictors--unexpected results

2016-01-27 Thread Richard L. Valliant
I'm been experimenting with the randomForest R package (v. 4.6-12) and getting an unexpected difference between rpart and randomForest results that may have something to do with using x's that are factors. The same model (see code below) is used to predict a 2-value variable called "resp" tha

Re: [R] randomForest set.seed()

2015-07-09 Thread Jeff Newmiller
Please post in plain text, and supply a reproducible example (that includes sample data, preferably using the dput function). The below code obtains repeatable random number sequences in each pass through the loop, as confirmed by the first value in each sequence. ## a <- 0 ## ## For loop to

[R] randomForest set.seed()

2015-07-09 Thread Joao Carreiras
Dear forum members, ​I wrote a piece of code to test various combinations of mtry and ntree, so that the best combination (in terms of mse​) could be used. Before each randomForest command I included a set.seed() command, so that I can keep track of the seed number and replicate the results. Howev

Re: [R] randomForest warning: The response has five or fewer unique values. Are you sure you want to do regression?

2014-03-25 Thread Sean Porter
667 Chlorophyll > # End   -Original Message- From: Liaw, Andy [mailto:andy_l...@merck.com] Sent: 25 March 2014 02:37 AM To: Sean Porter; r-help@r-project.org Subject: RE: [R] randomForest warning: The response has five or fewer unique values. Are you sure you want to do r

Re: [R] randomForest warning: The response has five or fewer unique values. Are you sure you want to do regression?

2014-03-24 Thread Liaw, Andy
it a factor. Best, Andy Liaw -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Sean Porter Sent: Thursday, March 20, 2014 3:27 AM To: r-help@r-project.org Subject: [R] randomForest warning: The response has five or fewer unique value

[R] randomForest warning: The response has five or fewer unique values. Are you sure you want to do regression?

2014-03-20 Thread Sean Porter
Hello everyone, Im relatively new to R and new to the randomForest package and have scoured the archives for help with no luck. I am trying to perform a regression on a set of predictors and response variables to determine the most important predictors. I have 100 response variables collected f

Re: [R] RandomForest, Party and Memory Management

2013-02-04 Thread Lorenzo Isella
Dear Dennis and dear All, It was probably not my best post. I am running R on a Debian box (amd64 architecture) and that is why I was surprised to see memory issues when dealing with a vector larger than 1Gb. The memory is there, but probably it is not contiguous. I will investigate into the matter

Re: [R] RandomForest, Party and Memory Management

2013-02-04 Thread Prof Brian Ripley
On Sun, 3 Feb 2013, Lorenzo Isella wrote: Dear All, For a data mining project, I am relying heavily on the RandomForest and Party packages. Due to the large size of the data set, I have often memory problems (in particular with the Party package; RandomForest seems to use less memory). I real

Re: [R] RandomForest, Party and Memory Management

2013-02-03 Thread Jeff Newmiller
Neither of your questions meets the Posting Guidelines (see footer of any email). 1) Not reproducible. [1] 2) Very operating-system specific and a FAQ. You have not indicated what your OS is (via sessionInfo), nor what reading you have done to address memory problems already (use a search engine

[R] RandomForest, Party and Memory Management

2013-02-03 Thread Lorenzo Isella
Dear All, For a data mining project, I am relying heavily on the RandomForest and Party packages. Due to the large size of the data set, I have often memory problems (in particular with the Party package; RandomForest seems to use less memory). I really have two questions at this point 1) Pl

Re: [R] RandomForest and Missing Values

2013-01-30 Thread nalluri pratap
If you wish to remove missing values, you can use the option na.action=na.omit.If you wish to Impute you can use rfImpute. --- On Mon, 28/1/13, Lorenzo Isella wrote: From: Lorenzo Isella Subject: [R] RandomForest and Missing Values To: r-h...@stat.math.ethz.ch Date: Monday, 28 January, 2013

[R] RandomForest and Missing Values

2013-01-28 Thread Lorenzo Isella
Dear All, I would like to use a randomForest algorithm on a dataset. The set is not particularly large/difficult to handle, but it has some missing values (both factors and numerical values). According to what I found https://stat.ethz.ch/pipermail/r-help/2005-September/078880.html https://stat.et

Re: [R] How do I make R randomForest model size smaller?

2012-12-04 Thread Liaw, Andy
uld be avoided with large datasets. Andy -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of John Foreman Sent: Monday, December 03, 2012 3:43 PM To: r-help@r-project.org Subject: [R] How do I make R randomForest model size smaller?

[R] How do I make R randomForest model size smaller?

2012-12-03 Thread John Foreman
I've been training randomForest models on 7 million rows of data (41 features). Here's an example call: myModel <- randomForest(RESPONSE~., data=mydata, ntree=50, maxnodes=30) I thought surely with only 50 trees and 30 terminal nodes that the memory footprint of "myModel" would be small. But it's

[R] randomForest Species Distribution Modelling

2012-06-06 Thread Fionn
Hi, I appologise if this is a rudimentary question and long winded but I just wanted to let ye know where I'm comming from. I'm new to R and I'm trying to use the 'randomForest' package to classify and predict. The Error message that is troubling me is: > pr<-predict(predictors,rf1, ext=ext) Error

Re: [R] randomForest Tutorial

2012-05-11 Thread Nagagautami Kodali
Hi Jim I am unable to access the webinar pdf on RandomForest Implementation. Can you please send it over? Regards -- Gautami This email message may contain proprietary, private and confidential information. The information transmitted is intended only for the

[R] randomForest

2012-02-08 Thread Anirudh Kondaveeti
Hi I have a dataset with a few variables but two variables are very important. One of them is say S1 and other S2. S1 is a numeric variable and S2 is a categorical variable. Consider two cases of S1-S2 pair like S1 = 70.2, S2 = A and S1 = 70.21, S2 = B. Assume my dataset has 20 instances of the f

Re: [R] RandomForest ... Always Returning a 1?

2012-02-05 Thread Anthony Presley
Peter, Thanks for checking on this! We're having some difficulties forecasting sales ... randomForest is doing a great job on days and periods where there is high variability. However, such as in this case, we do get dates/times/departments where it simply forecasts the most frequent value (I un

Re: [R] RandomForest ... Always Returning a 1?

2012-02-05 Thread Peter Langfelder
Well, if you look at your data more carefully, you will see that the histogram of y is heavily skewed towards 1 (small values). The 91/625 quantile is still 1 (there are 192 1s). It is therefore not surprising that RF comes up with mostly 1s (in my attempt it came up with two 2s, but that is a bit

[R] RandomForest ... Always Returning a 1?

2012-02-05 Thread apresley
We have a pretty simple R forecasting problem, using randomForest. Our script is below. We have data with about 6 columns, and 625 rows to do the training on, and then we're trying to forecast 90 or so rows. The x matrix has 3 years worth of sales data, starting at 6:15am. However, for some rea

Re: [R] randomForest: proximity for new objects using an existing rf

2012-02-01 Thread Liaw, Andy
ent: Wednesday, February 01, 2012 5:39 AM > To: r-help@r-project.org > Subject: [R] randomForest: proximity for new objects using an > existing rf > > Dear all, > > using an existing random forest, I would like to calculate > the proximity > for a new test object, i.e.

[R] randomForest: proximity for new objects using an existing rf

2012-02-01 Thread Kilian
Dear all, using an existing random forest, I would like to calculate the proximity for a new test object, i.e. the similarity between the new object and the old training objects which were used for building the random forest. I do not want to build a new random forest based on both old and new obj

[R] randomforest and AUC using 10 fold CV - Plotting results

2011-12-22 Thread David martin
Here is a snippet to show what i'm trying to do. library(randomForest) library(ROCR) library(caret) data(iris) iris <- iris[(iris$Species != "setosa"),] fit <- randomForest(factor(Species) ~ ., data=iris, ntree=50) train.predict <- predict(fit,iris,type="prob")[,2] plot(performance(prediction(t

Re: [R] randomForest - NaN in %IncMSE

2011-09-23 Thread Liaw, Andy
p-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of Katharine Miller > Sent: Tuesday, September 20, 2011 1:43 PM > To: r-help@r-project.org > Subject: [R] randomForest - NaN in %IncMSE > > Hi > > I am having a problem using varImpPlot in randomFo

[R] randomForest - NaN in %IncMSE

2011-09-20 Thread Katharine Miller
Hi I am having a problem using varImpPlot in randomForest. I get the error message "Error in plot.window(xlim = xlim, ylim = ylim, log = "") : need finite 'xlim' values" When print $importance, several variables have NaN under %IncMSE. There are no NaNs in the original data. Can someone hel

Re: [R] randomForest memory footprint

2011-09-08 Thread John Foreman
Message- >> From: r-help-boun...@r-project.org >> [mailto:r-help-boun...@r-project.org] On Behalf Of John Foreman >> Sent: Wednesday, September 07, 2011 2:46 PM >> To: r-help@r-project.org >> Subject: [R] randomForest memory footprint >> >> Hello, I am att

Re: [R] randomForest memory footprint

2011-09-08 Thread Liaw, Andy
an 10 terminal nodes per tree). Best, Andy > -Original Message- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of John Foreman > Sent: Wednesday, September 07, 2011 2:46 PM > To: r-help@r-project.org > Subject: [R] randomForest memory foo

[R] randomForest memory footprint

2011-09-07 Thread John Foreman
Hello, I am attempting to train a random forest model using the randomForest package on 500,000 rows and 8 columns (7 predictors, 1 response). The data set is the first block of data from the UCI Machine Learning Repo dataset "Record Linkage Comparison Patterns" with the slight modification that I

Re: [R] randomForest partial dependence plot variable names

2011-08-09 Thread Liaw, Andy
[i]), ylim=c(30, 70)) } par(op) Andy > -Original Message- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of Katharine Miller > Sent: Thursday, August 04, 2011 4:38 PM > To: r-help@r-project.org > Subject: [R] randomForest p

[R] randomForest partial dependence plot variable names

2011-08-04 Thread Katharine Miller
Hello, I am running randomForest models on a number of species. I would like to be able to automate the printing of dependence plots for the most important variables in each model, but I am unable to figure out how to enter the variable names into my code. I had originally thought to extract the

Re: [R] randomForest: too many elements specified?

2011-01-21 Thread Liaw, Andy
gt; From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of Czerminski, Ryszard > Sent: Thursday, January 20, 2011 1:08 PM > To: r-h...@stat.math.ethz.ch > Subject: [R] randomForest: too many elements specified? > > I getting "Error in matrix

[R] randomForest: too many elements specified?

2011-01-20 Thread Czerminski, Ryszard
rd.czermin...@astrazeneca.com RE: [R] randomForest: too many element specified? Liaw, Andy Mon, 17 Jan 2005 05:56:28 -0800 > From: luk > > When I run randonForest with a 169453x5 matrix, I got the > following message. > > Error in matrix(0, n, n) : matrix: too many elements specified > &g

Re: [R] randomForest speed improvements

2011-01-05 Thread Liaw, Andy
> > Sent: Tuesday, January 04, 2011 6:30 PM > > To: r-help@r-project.org > > Subject: Re: [R] randomForest speed improvements > > > > > > Andy, > > > > Thanks for the reply. I had no idea I could combine them > > back ... that > >

Re: [R] randomForest speed improvements

2011-01-05 Thread Liaw, Andy
egrading. Andy > -Original Message- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of apresley > Sent: Tuesday, January 04, 2011 6:30 PM > To: r-help@r-project.org > Subject: Re: [R] randomForest speed improvements > > > Andy, &

Re: [R] randomForest speed improvements

2011-01-04 Thread apresley
Andy, Thanks for the reply. I had no idea I could combine them back ... that actually will work pretty well. We can have several "worker threads" load up the RF's on different machines and/or cores, and then re-assemble them. RMPI might be an option down the road, but would be a bit of overhea

Re: [R] randomForest speed improvements

2011-01-04 Thread Liaw, Andy
ng only certain quantiles. The current RF code doesn't do this. Andy > -Original Message- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of apresley > Sent: Monday, January 03, 2011 6:28 PM > To: r-help@r-project.org > Subject: R

Re: [R] randomForest speed improvements

2011-01-03 Thread apresley
I haven't tried changing the mtry or ntree at all ... though I suppose with only 6 variables, and tens-of-thousands of rows, we can probably do less than 500 tree's (the default?). Although tossing the forest does speed things up a bit, seems to be about 15 - 20% faster in some cases, I need to k

Re: [R] randomForest speed improvements

2011-01-03 Thread Jonathan P Daily
ord... imbue it." - Jubal Early, Firefly r-help-boun...@r-project.org wrote on 01/03/2011 02:59:29 PM: > [image removed] > > [R] randomForest speed improvements > > apresley > > to: > > r-help > > 01/03/2011 03:03 PM > > Sent by: > > r-help

[R] randomForest speed improvements

2011-01-03 Thread apresley
Hi there, We're trying to use randomForest to do some predictions. The test-harness for our code is pretty straightforward: library ('randomForest'); data202 <- read.csv ("random.csv", header=TRUE); x<- data202[1:5,1:6]; y<- data202[1:5,8]; y<- y[,drop=TRUE]; x2 <- data202[

Re: [R] randomForest: tuneRF error

2010-12-23 Thread Uwe Ligges
Since we do not have the data, it is hard for us to reproduce and debug. Uwe Ligges On 21.12.2010 23:48, Dennis wrote: Just curious if anyone else has got this error before, and if so, would know what I could do (if anything) to get past it: mtry<- tuneRF(training, trainingdata$class, ntre

[R] randomForest: tuneRF error

2010-12-21 Thread Dennis
Just curious if anyone else has got this error before, and if so, would know what I could do (if anything) to get past it: > mtry <- tuneRF(training, trainingdata$class, ntreeTry = 500, stepFactor = 2, > improve = 0.05, trace = TRUE, plot = TRUE, doBest = FALSE) mtry = 13 OOB error = 0.62% Searc

Re: [R] randomForest: help with combine() function

2010-12-11 Thread Liaw, Andy
t; To: r-help@r-project.org > Subject: [R] randomForest: help with combine() function > > I've built two RF objects (RF1 and RF2) and have tried to combine > them, but I get the following error: > > Error in rf$votes + ifelse(is.na(rflist[[i]]$votes), 0, > rflist[[i]]

[R] randomForest: help with combine() function

2010-12-10 Thread Dennis Duro
I've built two RF objects (RF1 and RF2) and have tried to combine them, but I get the following error: Error in rf$votes + ifelse(is.na(rflist[[i]]$votes), 0, rflist[[i]]$votes) : non-conformable arrays In addition: Warning message: In rf$oob.times + rflist[[i]]$oob.times : longer object lengt

Re: [R] randomForest: How to append ID column along with predictions

2010-12-07 Thread Dennis Duro
- >> From: r-help-boun...@r-project.org >> [mailto:r-help-boun...@r-project.org] On Behalf Of Dennis Duro >> Sent: Tuesday, December 07, 2010 11:46 AM >> To: r-help@r-project.org >> Subject: [R] randomForest: How to append ID column along with >> prediction

Re: [R] randomForest: How to append ID column along with predictions

2010-12-07 Thread Liaw, Andy
..@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of Dennis Duro > Sent: Tuesday, December 07, 2010 11:46 AM > To: r-help@r-project.org > Subject: [R] randomForest: How to append ID column along with > predictions > > Hi all, > > When running a prediction

[R] randomForest: How to append ID column along with predictions

2010-12-07 Thread Dennis Duro
Hi all, When running a prediction using RF on another data, I get two columns returned: row number(?) and predicted class. Is there a way of appending the unique row value from an ID column in the dataframe to the predictions instead of the row number? I'm assuming that the returned results follow

[R] randomForest: How to append ID column along with predictions

2010-12-07 Thread Dennis
Hi all, When running a prediction using RF on another data, I get two columns returned: row number(?) and predicted class. Is there a way of appending the unique row value from an ID column in the dataframe to the predictions instead of the row number? I'm assuming that the returned results follow

Re: [R] randomForest parameters for image classification

2010-11-18 Thread Deschamps, Benjamin
the image in blocks and use predict.randomForest directly. Benjamin -Original Message- From: Liaw, Andy [mailto:andy_l...@merck.com] Sent: November 18, 2010 8:39 AM To: Deschamps, Benjamin; r-help@r-project.org Subject: RE: [R] randomForest parameters for image classification 1. Memory

Re: [R] randomForest parameters for image classification

2010-11-18 Thread Liaw, Andy
data you want to predict, not the other way around. Andy > -Original Message- > From: Deschamps, Benjamin [mailto:benjamin.descha...@agr.gc.ca] > Sent: Tuesday, November 16, 2010 11:16 AM > To: r-help@r-project.org > Cc: Liaw, Andy > Subject: RE: [R] randomForest pa

Re: [R] randomForest parameters for image classification

2010-11-16 Thread Deschamps, Benjamin
ks, Benjamin -Original Message- From: Liaw, Andy [mailto:andy_l...@merck.com] Sent: November 11, 2010 7:02 AM To: Deschamps, Benjamin; r-help@r-project.org Subject: RE: [R] randomForest parameters for image classification Please show us the code you used to run randomForest, the output, as well as wha

Re: [R] randomForest can not handle categorical predictors with more than 32 categories

2010-11-11 Thread Sven Garbade
You can try ctree in package party, but anyway: what is the deeper sense in a binary split for a variable with more than 32 levels? Regards, Sven 2010/11/10 Erik Iverson : > Well, the error message seems relatively straightforward. > > When you run str(x) (you did not provide the data) > > you sh

Re: [R] randomForest parameters for image classification

2010-11-11 Thread Liaw, Andy
p-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of Deschamps, Benjamin > Sent: Tuesday, November 09, 2010 10:52 AM > To: r-help@r-project.org > Subject: [R] randomForest parameters for image classification > > I am implementing an image classifica

Re: [R] randomForest can not handle categorical predictors with more than 32 categories

2010-11-10 Thread Erik Iverson
Well, the error message seems relatively straightforward. When you run str(x) (you did not provide the data) you should see 1 or more components are factors that have more than 32 levels. Apparently you can't include those predictors in a call to randomForest. You might find the following line

Re: [R] randomForest can not handle categorical predictors with more than 32 categories

2010-11-10 Thread Mattia Prosperi
try to transform the attributes that have more than 32 levels into dummy binary variables. 2010/11/10 Mai Dang : > I received this error > Error in randomForest.default(m, y, ...) : > Can not handle categorical predictors with more than 32 categories. > > using below code > > library(randomForest)

[R] randomForest can not handle categorical predictors with more than 32 categories

2010-11-10 Thread Mai Dang
I received this error Error in randomForest.default(m, y, ...) : Can not handle categorical predictors with more than 32 categories. using below code library(randomForest) library(MASS) memory.limit(size=12999) x <- read.csv("D:/train_store_title_view.csv", header=TRUE) x <- na.omit(x) set.seed(1

[R] randomForest parameters for image classification

2010-11-09 Thread Deschamps, Benjamin
I am implementing an image classification algorithm using the randomForest package. The training data consists of 31000+ training cases over 26 variables, plus one factor predictor variable (the training class). The main issue I am encountering is very low overall classification accuracy (a lot of

Re: [R] RandomForest Proximity Matrix

2010-10-21 Thread Liaw, Andy
From: Michael Lindgren > > Greetings R Users! > > I am posting to inquire about the proximity matrix in the randomForest > R-package. I am having difficulty pushing very large data through the > algorithm and it appears to hang on the building of the prox > matrix. I have > read on Dr. Breiman

[R] RandomForest Proximity Matrix

2010-10-21 Thread Michael Lindgren
Greetings R Users! I am posting to inquire about the proximity matrix in the randomForest R-package. I am having difficulty pushing very large data through the algorithm and it appears to hang on the building of the prox matrix. I have read on Dr. Breiman's website that in the original code a ch

Re: [R] randomForest - PartialPlot - reg

2010-09-24 Thread Vijayan Padmanabhan
Thanks for the elaborate detailing. I see sense now. Regards Vijayan Padmanabhan "What is expressed without proof can be denied without proof" - Euclide. "Liaw, Andy" 09/24/2010 04:31 PM To "Vijayan Padmanabhan" , "r-help" cc Subject RE: [R] r

Re: [R] randomForest - PartialPlot - reg

2010-09-24 Thread Liaw, Andy
help-boun...@r-project.org] On Behalf Of Vijayan Padmanabhan > Sent: Wednesday, September 22, 2010 11:47 PM > To: r-help > Subject: [R] randomForest - PartialPlot - reg > > > Dear R Group > I am not sure if this is the right forum to raise this query, > but i would > rather giv

[R] randomForest - PartialPlot - reg

2010-09-22 Thread Vijayan Padmanabhan
Dear R Group I am not sure if this is the right forum to raise this query, but i would rather give it a try and aim for reaching the right person who might be a part of this group who can help. I have a query on interpretation of PartialPlot in package randomForest. In my earlier queries in thi

Re: [R] randomForest - partialPlot - Reg

2010-09-22 Thread Liaw, Andy
> From: Vijayan Padmanabhan > > Dear R Group > I had an observation that in some cases, when I use the > randomForest model > to create partialPlot in R using the package "randomForest" > the y-axis displays values that are more than -1! > It is a classification problem that i was trying to addr

[R] randomForest - partialPlot - Reg

2010-09-21 Thread Vijayan Padmanabhan
Dear R Group I had an observation that in some cases, when I use the randomForest model to create partialPlot in R using the package "randomForest" the y-axis displays values that are more than -1! It is a classification problem that i was trying to address. Any insights as to how the y axis can

Re: [R] randomForest outlier return NA

2010-07-15 Thread Pau Carrio Gaspar
.org [mailto:r-help-boun...@r-project.org] > On Behalf Of Pau Carrio Gaspar > Sent: Wednesday, July 14, 2010 6:36 AM > To: r-help@r-project.org > Subject: [R] randomForest outlier return NA > > Dear R-users, > > I have a problem with randomForest{outlier}. > Afte

Re: [R] randomForest outlier return NA

2010-07-15 Thread Liaw, Andy
anks for reporting. Best, Andy -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Pau Carrio Gaspar Sent: Wednesday, July 14, 2010 6:36 AM To: r-help@r-project.org Subject: [R] randomForest outlier return NA Dear R-users, I have a problem

[R] randomForest outlier return NA

2010-07-14 Thread Pau Carrio Gaspar
Dear R-users, I have a problem with randomForest{outlier}. After running the following code ( that produces a silly data set and builds a model with randomForest ): ### library(randomForest) set.seed(0) ## build data set X <- rbind( matrix( runif(n=400,min=-1,max=1), ncol =

Re: [R] randomForest: sampling with replacement?

2010-05-10 Thread Max Kuhn
>> Would you please confirm that it is a bootstrap sample with >> replacement? Someone should note that the definition of a bootstrap sample is a sample with replacement (usually of size n). I've read quite a few papers where they claim to be using the bootstrap. Upon further review (sometimes to

Re: [R] randomForest: sampling with replacement?

2010-05-10 Thread Liaw, Andy
See the "replace" argument in ?randomForest. Andy > -Original Message- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of Dimitri > Liakhovitski > Sent: Friday, May 07, 2010 12:21 PM > To: r-help > Subject:

[R] randomForest: sampling with replacement?

2010-05-07 Thread Dimitri Liakhovitski
Hello! I know that as trees are constructed, the root node contains a bootstrap sample of data of the same size as original the data set. Would you please confirm that it is a bootstrap sample with replacement? I think it is - just wanted to make sure. Thank you very much! -- Dimitri Liakhovitsk

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Dimitri Liakhovitski
>> Andy, I'll explain why I am asking. I probably should have >> done it in the beginning: >> I am asking not in order to figure out how to do it. I am >> asking in order to figure something that' was done around >> November 01, 2008. >> Back then, a piece of code was run where from the object of >

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Liaw, Andy
From: Dimitri Liakhovitski > >> Andy, I'll explain why I am asking. I probably should have > done it in > >> the beginning: > >> I am asking not in order to figure out how to do it. I am > asking in > >> order to figure something that' was done around November 01, 2008. > >> Back then, a piece

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Dimitri Liakhovitski
Thank you very much, Andy. I did turn off HTML - hope it'll solve the problem! > Andy, but it is the FIRST column in $importance (not the SECOND) that is > labeled "%IncMSE". The second column is labeled "IncNodePurity". So, I > am confused - which one is the mean decrease in accuracy? > Or, maybe

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Liaw, Andy
> From: Dimitri Liakhovitski > Thank you very much, Andy. > I did turn off HTML - hope it'll solve the problem! Indeed it does! [...] > > [AL]: As I said, you are recommended to use importance() to extract > > variable importance.  The recommendation is for avoiding confusions > > like yours

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Liaw, Andy
Not that I want to pick on you, but can you turn off the html format in your messages? The mailing list balk at such format, and I can't reply in plain text with the right formatting of previous messages (had to manually remove the tabbed indents that Outlook added when changed to plain text). Se

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Dimitri Liakhovitski
Andy, thank you - and sorry for being a bit slow (see my questions below): On Thu, May 6, 2010 at 8:37 AM, Liaw, Andy wrote: > See reply inline below. > > Andy > > From: Dimitri Liakhovitski > > > > I have a question about predictor importances in randomForest. > > > > Once I've run randomForest

Re: [R] randomForest: predictor importance (for regressions)

2010-05-06 Thread Liaw, Andy
See reply inline below. Andy From: Dimitri Liakhovitski > > I have a question about predictor importances in randomForest. > > Once I've run randomForest and got my object, I get their importances: > rfresult$importance > I also get the "standard errors" of the permutation-based importance > m

[R] randomForest: predictor importance (for regressions)

2010-05-05 Thread Dimitri Liakhovitski
I have a question about predictor importances in randomForest. Once I've run randomForest and got my object, I get their importances: rfresult$importance I also get the "standard errors" of the permutation-based importance measure: rfresult$importanceSD I have 2 questions: 1. Because I am dealin

[R] RandomForest diagnostics plot

2010-04-29 Thread Changbin Du
HI, Andy, ON the RandomForest diagnostics plot, three lines were drawn. ON the help document of plot.randomForest, only the error rate and MSE are described. I dont know which line is which? Can you help me with that? Thanks so much! -- Sincerely, Changbin -- [[alternative HTML ver

Re: [R] randomForest predictions with new data

2010-04-25 Thread Uwe Ligges
On 25.04.2010 16:27, Rolf Edberg wrote: Hi I am new to R, randomForest and I have read about how to use it in your old mails. I have also run the predictions examples from CRAN. But I still don't understand how to use it right. The thing that I don't understand is how to run

[R] randomForest predictions with new data

2010-04-25 Thread Rolf Edberg
Hi I am new to R, randomForest and I have read about how to use it in your old mails. I have also run the predictions examples from CRAN. But I still don't understand how to use it right. The thing that I don't understand is how to run the result from the randomForest on one

Re: [R] RandomForest how to identify two classes when only one is present

2010-04-08 Thread Steve Lianoglou
Hi, On Thu, Apr 8, 2010 at 1:18 PM, JMark wrote: > > I'm trying to do: > randomForest(f, data = moths.train) > > But I get this error: > Error in randomForest.default(m, y, ...) : >  Need at least two classes to do classification. > > When I look at the data for this, I realize there are no posit

[R] RandomForest how to identify two classes when only one is present

2010-04-08 Thread JMark
I'm trying to do: randomForest(f, data = moths.train) But I get this error: Error in randomForest.default(m, y, ...) : Need at least two classes to do classification. When I look at the data for this, I realize there are no positive cases of this item: [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Re: [R] randomForest maxnodes

2010-01-15 Thread Liaw, Andy
.@r-project.org] On Behalf Of Epic John > Sent: Friday, January 15, 2010 4:50 PM > To: r-help@r-project.org > Subject: [R] randomForest maxnodes > > Has anyone sucessfully used the maxnodes feature in > randomForest? I tried > setting it, but when it is non-NULL I always get

[R] randomForest maxnodes

2010-01-15 Thread Epic John
Has anyone sucessfully used the maxnodes feature in randomForest? I tried setting it, but when it is non-NULL I always get back a forest in which all trees have size 1. I am using a continuous response (regression). Any help would be appreciated. Thanks. [[alternative HTML version deleted]

Re: [R] RandomForest - getTree status code

2009-12-07 Thread NCS
. ncs - Original Message From: "Liaw, Andy" To: NCS ; r-help@r-project.org Sent: Mon, December 7, 2009 5:46:24 AM Subject: RE: [R] RandomForest - getTree status code Is that the entire tree? If so there's a problem. The node status is defined as follows in rf.h of

Re: [R] RandomForest - getTree status code

2009-12-07 Thread Liaw, Andy
--- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] On Behalf Of NCS > Sent: Friday, December 04, 2009 4:53 PM > To: r-help@r-project.org > Subject: [R] RandomForest - getTree status code > > > What does a status value of -3 mean when I do

[R] RandomForest - getTree status code

2009-12-04 Thread NCS
What does a status value of -3 mean when I do a regression with RF and use the getTree function? left daughter right daughter split var split point status prediction 12 311 4.721000e+03 -3 15.8489576 24 5 5 6.500

[R] randomForest bug in predict.randomForest()

2009-11-23 Thread J Sousa
The NEWS of the randomForest R library mention that version 4.5-13 fixed a bug in predict.randomForest() when newdata is a matrix with no rownames. Can anyone explain what were the consequences of this bug in the predictions? I think the bug was fixed with the new line if (is.null(rn)) rn <- kee

[R] randomForest: impact of bug fixed in version 4.5-13

2009-11-19 Thread J Sousa
The NEWS of the randomForest R library mention that version 4.5-13 fixed a bug in predict.randomForest() when newdata is a matrix with no rownames. I think it corresponds to the difference in files predict.randomForest.R which is the new line if (is.null(rn)) rn <- keep As I've been using version

Re: [R] RandomForest & PMML

2009-11-04 Thread NCS
Thanks for the feedback. In the meantime, is there any other way to utilize the model outside of R, even if it is a little more complicated? - Original Message From: Graham Williams To: NCS Cc: r-help@r-project.org Sent: Tue, November 3, 2009 9:53:10 PM Subject: Re: [R

Re: [R] RandomForest & PMML

2009-11-03 Thread Graham Williams
2009/11/4 NCS : > I cannot seem to write a randomforest model in PMML - either through calling > PMML(model) or through Rattle.  It appears that it is not yet supported. > Randomsurvivalforest is, but not randomforest.  Any ideas on possible > workarounds for this? > > Thanks > ncs It is not yet i

  1   2   >