I am getting an error, while running the below code:
library(neuralnet)
library(datasets.load)
Neural_Net = neuralnet(formula = dist ~ mag , data = attenu, hidden = C(1,3) )
I get the below error:
Error in C(1, 3) : object not interpretable as a factor
I couldn't solve the problem. Where am I doing wrong?
I will be very glad for any help. Thanks a lot.
Related
guys! I encountered an error when creating a cost surface based on DEM data.
Adm1 <-
rgeoboundaries::geoboundaries(
country = "Austria",
adm_lvl = "adm0")
r <- get_elev_raster(Adm1,z=4)
cs <- create_slope_cs(dem = r, cost_function = 'tobler', neighbours = 16)
The error is like
Error in h(simpleError(msg, call)) :
error in evaluating the argument 'x' in selecting a method for function 'extent': ‘srs’ is not a slot in class “BasicRaster”
I can run this code on my other laptop. So I think there's nothing wrong with my code. But do you know how to fix this?
I am using the airbnb dataset. After cleaning it, I tried to apply a random forest (I did a tree and a pruned tree and they worked). I don't have a lot of experience but here is my code :
split_index <- createDataPartition(airbnbcleanedfinal$logprice, p = 0.8, list = F)
#Use index to split data
training<-training <- airbnbcleanedfinal[split_index,]
training1 <- airbnbcleanedfinal[sample(nrow(airbnbcleanedfinal),100000,replace=TRUE),]
features_test <- airbnbcleanedfinal[-split_index, !(colnames(airbnbcleanedfinal) %in% c('logprice'))]
target_test <- airbnbcleanedfinal[-split_index, 'logprice']
library(randomForest)
rf_train <- randomForest(logprice ~ ., data = airbnbcleanedfinal,
subset=training,
mtry = 5)
But I always get the same error message :
Error in xj[i] : invalid subscript type 'list'
I also tried to delete subset=training and put directly data=training but it makes R run forever. I also tried using training1 that I created for that purpose but still got the same error message.
I tried unlist(training) but it did not work. My data is huge (85k-15 variables) too, maybe that is the problem? How can I force training to be a list?
I am trying to run a supervised SOM model based on cross-validated values.
The issue seems is in the sup.som line. I receive this error: Error in !toroidal : invalid argument type.
This code has worked fine in the past (within past few days) and I have tried restarting RStudio. There was a very similar question here (Error in !toroidal : invalid argument type in R with som package) but the issue seemed to have resolved itself for the other person. Any help is appreciated!
My code is:
require(kohonen)
set.seed(123)
fitControl <- trainControl(method = "cv",number = 10)
tg <- expand.grid(xdim=c(3:10), ydim=c(3:10), user.weights=seq(0.1,0.9,by=0.1), topo="hexagonal")
somFit1 <- train(train[,1:6], as.factor(train$CORR), method="xyf", trControl=fitControl, tuneLength=20, tuneGrid=tg)
'''sup.som <- xyf(training.sc, classvec2classmat(corrupt.train), grid = somgrid(xdim=somFit1$bestTune$xdim, ydim=somFit1$bestTune$ydim, topo="hexagonal"), user.weights=somFit1$bestTune$user.weights, keep.data=TRUE)
I ended up trying my code in the R console to see if I could get more detailed error messages, and I got the following:
require(class)
Loading required package: class
Attaching package: ‘class’
The following object is masked from ‘package:kohonen’:
somgrid
The solution was to specify kohonen::somgrid before my code as follows:
sup.som <- xyf(training.sc, classvec2classmat(corrupt.train), grid = kohonen::somgrid(xdim=somFit1$bestTune$xdim, ydim=somFit1$bestTune$ydim, topo="hexagonal"), user.weights=somFit1$bestTune$user.weights, keep.data=TRUE)
Hope this helps someone else
I ran into a problem with the DEoptim package. I am trying to minimize the function minF by optimizing vector optVector (see code).
##construct function
minF <- function(x, y, z){
return(mean((z-rowSums(t(x*t(y))))^2))
}
#random matrix and vector
testmrx <- matrix(rnorm(6),38,9)
vctr <- runif(38, min=0, max=50)
#Vector to be optimized and its bounds
optVector = c(20,20,50,30,30,10,3,5,5)
lowr = c(0,0,0,0,0,0,0,0,0)
uppr = c(50,50,200,100,100,50,20,20,20)
#Call of DEoptim
DEoptim(fn = minF(optVector, testmrx, vctr), lower=lowr, upper=uppr)
When I try to do this I get following error:
Error in get(as.character(FUN), mode = "function", envir = envir) :
object 'fn' of mode 'function' was not found
I found a similar error posted (link), but the proposed solution of changing variable names did not work. I have no idea what might be causing the problem here. Any help would be greatly appreciated!
The moment condition function is simply exp(-g/r)-1, where g is a known series of daily return on AAA-class bond index, and r is the rikiness measure to be derived through gmm. My codes are as follows:
View(Source)
library(gmm)
data(Source)
x <- Source[1:5200,"AAA"]
m <- function(r,x)
{m.1 <- exp(-x[,"AAA"]/r)-1}
summary(gmm(m,x,t0=1,method="BFGS",control=1e-12))
Which in term yields the following error message:
****Error in model.frame.default(formula = gmat ~ 1, drop.unused.levels = TRUE) :
invalid type (list) for variable 'gmat'****
Could anyone help me figure out what went wrong?
Thanks a lot!
For those kind people who would like to replicate the results, please find attached the source data as mentioned above.
The correct r is 1.590 , which can be solved through goal searching in excel, with target function :(average(exp(-g/r)-1) )^2 and target value: 0 (tolerance: 1e-12)
https://docs.google.com/spreadsheets/d/1AnTErQd2jm9ttKDZa7On3DLzEZUWaz5Km3nKaB7K18o/edit?usp=sharing