Error using colSums on list or data frame

Asked Jun 06 '18 at 14:18

Active Jun 16 '18 at 17:01

Viewed 115 times

I am trying to use a simple ifelse command to do a simple operation on a dataframe.

mnist_raw <- read.csv("https://pjreddie.com/media/files/mnist_train.csv")
mnist = head(mnist_raw,10000)
H = mnist[1:1000,250:300]
H1 = ifelse(H>0,H,1)


colSums(H)

#Whereas this gives an error. Error in colSums(H1) : 'x' must be an array of at least two dimension
colSums(H1)

It seems that the output of ifelse is a list, which is a different type from the input. Any ideas on how to tackle this so that I get the same type of output as the input.

edited Jun 16 '18 at 17:01

camille

16,432
18
38
60

asked Jun 06 '18 at 14:18

honeybadger

1,465
1
19
32

1

related: https://stackoverflow.com/questions/9449184/if-else-vs-ifelse-with-lists – pogibas Jun 06 '18 at 14:22
5

`read.csv` creates a dataframe, so `H` will be a dataframe, which is a special type of list. So the output is the same type as the input. Perhaps you want `H <- as.matrix(H)`. – Andrew Gustar Jun 06 '18 at 14:24
@ PoGibas. Thank you! – honeybadger Jun 06 '18 at 14:26
@AndrewGustar: That solved my problem. Thank you! – honeybadger Jun 06 '18 at 14:28

Error using colSums on list or data frame

0 Answers0