Hello,
I've been experimenting with the H2O package, which seems to be a very
interesting and promising project.
I'm getting a few error messages through using "h2o.deeplearning" (which I
guess it must be something I'm doing wrong). Here is a reproducible example
of errors using the "n_folds" argument to this function and also when
supplying a list to the argument "hidden" in an attempt to perform
parameter search.
BTW - I'm running H20 on my local host (Mac/OS).
library(h2o)
localH2O <- h2o.init()
iris.hex <- as.h2o(iris)
iris.dl <- h2o.deeplearning(x = 1:4, y = 5, training_frame = iris.hex)
#n_folds issue
iris.dl2 <- h2o.deeplearning(x = 1:4, y = 5, n_folds = 5, training_frame =
iris.hex)
Error in .h2o.doSafeREST(conn = conn, h2oRestApiVersion =
h2oRestApiVersion, :
Unknown parameter: n_folds
>
#Parameter search issue
iris.dl3 <- h2o.deeplearning(x = 1:4, y = 5, hidden=list(c(5,5), c(10,10)),
training_frame = iris.hex)
Error in which(params[[i$name]] == Inf | params[[i$name]] == -Inf) :
(list) object cannot be coerced to type 'double'
Any pointers would be much appreciated.
Thanks,
Axel.
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
