I'm working on a Regression problem with Neural Networks using H2O.
My dataset is really big with millions of observations and lot of continuous and discreate variables.
When initializing H2O I do:
h2o.init(ip = "<IP>", port = <PORT>, ...)
I'm wondereing if is there any chance for me to train my models on remote computers. Of course, I know I will have to pay for that service.
Any suggestion about this?
Thanks!