Try to run our code on a bigger machine

Currently, we get a MemoryError during preprocessing the full training data. Could a bigger machine solve this issue?

https://cloud.google.com/compute/docs/machine-types
https://cloud.google.com/sdk/gcloud/reference/compute/instances/set-machine-type