CODEWITHSUNDEEP
r
bernie2436
bernie2436

Reputation: 23921

Why doesn't predict return any rows or columns in R?

I am very new to R. I have data that looks like this: 

> head(NB)
       a     s    e    i
9011 20-30   F  Others 10-50K
9012 GT 45   M  Others 10-50K

I classify it with naiveBayes like this: 

c = i ~ a + s + e
cl = naiveBayes(c, head(NB,1500), laplace = 0)

Then I predict its outcome on the new data like this 

> p <- predict(classifier, tail(NB, 500), type = c("class", "raw"), threshold = 0.001)

I want to look at the prediction for each datapoint in p and see how well it matches up with the actual value for p -- but I can't figure out what p actually represents. It seems to have no rows and no columns -- but it plots into a histogram that seems to show predictions from the data. 

> nrow(p)
NULL
> ncol(p)
NULL
> str(p) says 
Factor w/ 3 levels "10-50K","50-80K",..: 1 1 1 1 1 1 1 1 1 1 ...

What is going on? How do I find out what it predicts, for say, the 3rd value in the P dataset? Why doesn't p have any rows or columns?

Upvotes: 0

Views: 131

Answers (1)

user2054817
user2054817

Reputation: 88

p is a vector of factors. In R, vectors do not have a number of rows or columns, only a length. Typing length(p) will give you the length. Each element of p is one of "10-50K", "50-80K", or a third value. To see the different values in p, type unique(p). To get the third element of p, just access it as you would with any other vector p[3] or to see all of p print(p). If you want to count the number that are the same as your original data, try sum(p == NB$i). Have a look here for more info http://www-users.cs.york.ac.uk/~jc/teaching/arin/R_practical/.

Upvotes: 1

Related Questions