Error while using predict function for h2o.deeplearning model on raster stack
I am trying to use `h2o.deeplearning` model to predict on raster data. It returns me the following error
> Error: Not compatible with requested type: [type=character; target=double].
Here is a minimal, reproducible, self-contained example
```
library(terra)
library(h2o)
library(tidyverse)
h2o.init()
# create a RasterStack or RasterBrick with with a set of predictor layers
logo <- rast(system.file("external/rlogo.grd", package="raster"))
names(logo)
# known presence and absence points
p <- matrix(c(48, 48, 48, 53, 50, 46, 54, 70, 84, 85, 74, 84, 95, 85,
66, 42, 26, 4, 19, 17, 7, 14, 26, 29, 39, 45, 51, 56, 46, 38, 31,
22, 34, 60, 70, 73, 63, 46, 43, 28), ncol=2)
a <- matrix(c(22, 33, 64, 85, 92, 94, 59, 27, 30, 64, 60, 33, 31, 9,
99, 67, 15, 5, 4, 30, 8, 37, 42, 27, 19, 69, 60, 73, 3, 5, 21,
37, 52, 70, 74, 9, 13, 4, 17, 47), ncol=2)
# extract values for points
xy <- rbind(cbind(1, p), cbind(0, a))
v <- data.frame(cbind(pa=xy[,1], terra::extract(logo, xy[,2:3]))) %>%
mutate(pa = as.factor(pa))
str(v)
#### Import data to H2O cluster
df <- as.h2o(v)
#### Split data into train, validation and test dataset
splits <- h2o.splitFrame(df, c(0.70,0.15), seed=1234)
train <- h2o.assign(splits[[1]], "train.hex")
valid <- h2o.assign(splits[[2]], "valid.hex")
test <- h2o.assign(splits[[3]], "test.hex")
#### Create response and features data sets
y <- "pa"
x <- setdiff(names(train), y)
### Deep Learning Model
dl_model <- h2o.deeplearning(training_frame=train,
validation_frame=valid,
x=x,
y=y,
standardize=TRUE,
seed=125)
dnn_pred <- function(model, data, ...) {
predict(model, newdata=as.h2o(data), ...)
}
p <- predict(logo, model=dl_model, fun=dnn_pred)
plot(p)
```
