Tuning hyper-parameters using grid searches in H2O

H2O packages also allow you to perform hyper-parameter tuning using grid search (h2o.grid).

Getting ready

We first load and initialize the H2O package with the following code:

# Load the required packages
require(h2o)

# Initialize H2O instance (single node)
localH2O = h2o.init(ip = "localhost", port = 54321, startH2O = TRUE,min_mem_size = "20G",nthreads = 8)

The occupancy dataset is loaded, converted to hex format, and named occupancy_train.hex.

How to do it...

The section will focus on optimizing hyper parameters in H2O using grid searches.

1. In our case, we will optimize for the activation function, the number of hidden layers (along with the number of neurons in each layer), epochs, and regularization lambda (l1 and l2):

# Perform hyper parameter tuning
activation_opt <- c("Rectifier","RectifierWithDropout", "Maxout","MaxoutWithDropout")
hidden_opt <- list(5, c(5,5))
epoch_opt <- c(10,50,100)
l1_opt <- c(0,1e-3,1e-4)
l2_opt <- c...